docs(*): updated CI pipeline name and added .gitattributes

Signed-off-by: k4yt3x <i@k4yt3x.com>

.gitattributes

@@ -0,0 +1 @@
+models/* linguist-vendored

.github/FUNDING.yml

@@ -1,6 +1,6 @@
 # These are supported funding model platforms
 
-github: # Replace with up to 4 GitHub Sponsors-enabled usernames e.g., [user1, user2]
+github: k4yt3x
 patreon: k4yt3x
 open_collective: # Replace with a single Open Collective username
 ko_fi: # Replace with a single Ko-fi username

.github/ISSUE_TEMPLATE/bug-report.md

@@ -1,32 +0,0 @@
----
-name: Bug report
-about: Use this template if you believe you've encountered a bug.
-title: ''
-labels: bug
-assignees: K4YT3X
-
----
-
-## Environment Information
-
-|Module|Version|
-|-|-|
-|`Video2X`||
-|`FFmpeg`||
-|`waifu2x-caffe`||
-|`waifu2x-converter-cpp`||
-|`waifu2x-ncnn-vulkan`||
-
-## Symptom
-
-Briefly describe what's going on.
-
-## Screenshots or Error Messages
-
-Please paste screenshots or error messages here.
-Include as much details as you can, as this greatly helps debugging the problem.
-
-Please paste your error message in the code block down below:
-
-```
-```

.github/ISSUE_TEMPLATE/feature_request.md

@@ -1,16 +0,0 @@
----
-name: Feature request
-about: Request a new feature to be added
-title: ''
-labels: enhancement
-assignees: K4YT3X
-
----
-
-## Description
-
-Briefly describe the feature you want to be added into `video2x`.
-
-## Sources
-
-Paste links to descriptions of related documentations, websites and etc. here.

.github/workflows/build.yml

@@ -0,0 +1,82 @@
+name: build
+on:
+  push:
+    branches:
+      - master
+      - dev
+  pull_request: {}
+  workflow_dispatch: {}
+jobs:
+  ubuntu:
+    runs-on: ubuntu-latest
+    env:
+      DEBIAN_FRONTEND: noninteractive
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          submodules: recursive
+      - name: Install dependencies
+        run: |
+          git submodule update --init --recursive
+          sudo apt-get update
+          sudo apt-get install -y --no-install-recommends \
+            libavcodec-dev \
+            libavdevice-dev \
+            libavfilter-dev \
+            libavformat-dev \
+            libavutil-dev \
+            libswscale-dev \
+            libvulkan-dev \
+            glslang-tools \
+            libomp-dev
+      - name: Build Video2X
+        run: |
+          mkdir -p /tmp/build /tmp/install
+          cmake -B /tmp/build -S . -DUSE_SYSTEM_NCNN=OFF \
+            -DCMAKE_C_COMPILER=clang -DCMAKE_CXX_COMPILER=clang++ \
+            -DCMAKE_BUILD_TYPE=Debug -DCMAKE_INSTALL_PREFIX=/tmp/install \
+            -DINSTALL_BIN_DESTINATION=. -DINSTALL_INCLUDE_DESTINATION=include \
+            -DINSTALL_LIB_DESTINATION=. -DINSTALL_MODEL_DESTINATION=.
+          cmake --build /tmp/build --config Debug --target install --parallel
+      - name: Upload artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: video2x-nightly-linux-amd64
+          path: /tmp/install
+  windows:
+    runs-on: windows-latest
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          submodules: recursive
+      - name: Install Vulkan SDK
+        uses: humbletim/setup-vulkan-sdk@v1.2.0
+        with:
+          vulkan-query-version: 1.3.204.0
+          vulkan-components: Vulkan-Headers, Vulkan-Loader, Glslang, SPIRV-Tools, SPIRV-Headers
+      - name: Install dependencies
+        shell: pwsh
+        run: |
+          $ffmpegVersion = "7.1"
+          $ncnnVersion = "20240820"
+
+          git submodule update --init --recursive
+
+          curl -Lo ffmpeg-shared.zip "https://github.com/GyanD/codexffmpeg/releases/download/$ffmpegVersion/ffmpeg-$ffmpegVersion-full_build-shared.zip"
+          Expand-Archive -Path ffmpeg-shared.zip -DestinationPath third_party
+          Rename-Item -Path "third_party/ffmpeg-$ffmpegVersion-full_build-shared" -NewName ffmpeg-shared
+
+          curl -Lo ncnn-shared.zip "https://github.com/Tencent/ncnn/releases/download/$ncnnVersion/ncnn-$ncnnVersion-windows-vs2022-shared.zip"
+          Expand-Archive -Path ncnn-shared.zip -DestinationPath third_party
+          Rename-Item -Path "third_party/ncnn-$ncnnVersion-windows-vs2022-shared" -NewName ncnn-shared
+      - name: Build Video2X
+        run: |
+          cmake -S . -B build -DUSE_SYSTEM_NCNN=OFF -DCMAKE_BUILD_TYPE=Debug -DCMAKE_INSTALL_PREFIX=build/video2x_install
+          cmake --build build --config Debug --parallel --target install
+      - name: Upload artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: video2x-nightly-windows-amd64
+          path: build/video2x_install

.github/workflows/release.yml

@@ -0,0 +1,55 @@
+name: Release
+on:
+  push:
+    tags:
+      - "*"
+
+jobs:
+  setup:
+    if: github.event.base_ref == 'refs/heads/master'
+    name: Setup
+    runs-on: ubuntu-latest
+    outputs:
+      tag: ${{ steps.get_tag.outputs.tag }}
+    steps:
+      - name: Get tag
+        id: get_tag
+        run: echo ::set-output name=tag::${GITHUB_REF/refs\/tags\//}
+
+  container:
+    name: Build and upload container
+    needs:
+      - setup
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+      - uses: mr-smithers-excellent/docker-build-push@v5
+        name: Build & push the Docker image
+        with:
+          registry: ghcr.io
+          username: ${{ secrets.GHCR_USER }}
+          password: ${{ secrets.GHCR_TOKEN }}
+          dockerfile: Dockerfile
+          image: video2x
+          tags: latest, ${{ needs.setup.outputs.tag }}
+
+  create-release:
+    name: Create release
+    needs:
+      - setup
+      - container
+    runs-on: ubuntu-latest
+    outputs:
+      upload_url: ${{ steps.create_release.outputs.upload_url }}
+    steps:
+      - name: Create release
+        id: create_release
+        uses: actions/create-release@v1
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        with:
+          tag_name: ${{ needs.setup.outputs.tag }}
+          release_name: Video2X ${{ needs.setup.outputs.tag }}
+          draft: true
+          prerelease: false

.gitignore

@@ -1,107 +1,77 @@
-# PyCharm
-.idea/
+# Data files
+data/
 
-# Byte-compiled / optimized / DLL files
-__pycache__/
-*.py[cod]
-*$py.class
-
-# C extensions
-*.so
-
-# Distribution / packaging
-.Python
+# Built-binaries
+bin/
 build/
-develop-eggs/
-dist/
-downloads/
-eggs/
-.eggs/
-lib/
-lib64/
-parts/
-sdist/
-var/
-wheels/
-*.egg-info/
-.installed.cfg
-*.egg
-MANIFEST
 
-# PyInstaller
-#  Usually these files are written by a python script from a template
-#  before PyInstaller builds the exe, so as to inject date/other infos into it.
-*.manifest
-*.spec
+# Prerequisites
+*.d
 
-# Installer logs
-pip-log.txt
-pip-delete-this-directory.txt
+# Object files
+*.o
+*.ko
+*.obj
+*.elf
 
-# Unit test / coverage reports
-htmlcov/
-.tox/
-.coverage
-.coverage.*
-.cache
-nosetests.xml
-coverage.xml
-*.cover
-.hypothesis/
-.pytest_cache/
+# Linker output
+*.ilk
+*.map
+*.exp
 
-# Translations
-#*.mo
-#*.pot
+# Precompiled Headers
+*.gch
+*.pch
 
-# Django stuff:
-*.log
-local_settings.py
-db.sqlite3
+# Libraries
+*.lib
+*.a
+*.la
+*.lo
 
-# Flask stuff:
-instance/
-.webassets-cache
+# Shared objects (inc. Windows DLLs)
+*.dll
+#*.so
+*.so.*
+*.dylib
 
-# Scrapy stuff:
-.scrapy
+# Executables
+*.exe
+*.out
+*.app
+*.i*86
+*.x86_64
+*.hex
 
-# Sphinx documentation
-docs/_build/
+# Debug files
+*.dSYM/
+*.su
+*.idb
+*.pdb
 
-# PyBuilder
-target/
+# Kernel Module Compile Results
+#*.mod*
+*.cmd
+.tmp_versions/
+modules.order
+Module.symvers
+Mkfile.old
+dkms.conf
 
-# Jupyter Notebook
-.ipynb_checkpoints
+# clang tooling compilation databases
+.cache/
+*.plist
 
-# pyenv
-.python-version
-
-# celery beat schedule file
-celerybeat-schedule
-
-# SageMath parsed files
-*.sage.py
-
-# Environments
-.env
-.venv
-env/
-venv/
-ENV/
-env.bak/
-venv.bak/
-
-# Spyder project settings
-.spyderproject
-.spyproject
-
-# Rope project settings
-.ropeproject
-
-# mkdocs documentation
-/site
-
-# mypy
-.mypy_cache/
+# CMake
+CMakeLists.txt.user
+CMakeCache.txt
+CMakeFiles
+CMakeScripts
+Testing
+Makefile
+cmake_install.cmake
+install_manifest.txt
+compile_commands.json
+CTestTestfile.cmake
+_deps
+CMakeUserPresets.json

.gitmodules

@@ -0,0 +1,6 @@
+[submodule "third_party/libreal_esrgan_ncnn_vulkan"]
+	path = third_party/libreal_esrgan_ncnn_vulkan
+	url = https://github.com/k4yt3x/libreal-esrgan-ncnn-vulkan.git
+[submodule "third_party/ncnn"]
+	path = third_party/ncnn
+	url = https://github.com/Tencent/ncnn.git

CHANGELOG.md

@@ -0,0 +1,8 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]

CMakeLists.txt

@@ -0,0 +1,338 @@
+cmake_minimum_required(VERSION 3.10)
+project(video2x VERSION 6.0.0 LANGUAGES CXX C)
+
+# Set the C standard
+set(CMAKE_C_STANDARD 99)
+set(CMAKE_C_STANDARD_REQUIRED ON)
+
+# Set the C++ standard
+set(CMAKE_CXX_STANDARD 17)
+set(CMAKE_CXX_STANDARD_REQUIRED ON)
+set(CMAKE_CXX_EXTENSIONS OFF)
+
+# Set the default build type to Release if not specified
+if(NOT CMAKE_BUILD_TYPE)
+    set(CMAKE_BUILD_TYPE Release)
+endif()
+
+# Set the default optimization flags for Release builds
+if(CMAKE_BUILD_TYPE STREQUAL "Release")
+    if (CMAKE_CXX_COMPILER_ID STREQUAL "MSVC")
+        set(CMAKE_C_FLAGS_RELEASE "${CMAKE_C_FLAGS_RELEASE} /Ox /GL /LTCG /MD /DNDEBUG")
+        set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} /Ox /GL /LTCG /MD /DNDEBUG")
+    elseif (CMAKE_CXX_COMPILER_ID STREQUAL "GNU" OR CMAKE_CXX_COMPILER_ID MATCHES "Clang")
+        set(CMAKE_C_FLAGS_RELEASE "${CMAKE_C_FLAGS_RELEASE} -O3 -march=native -flto")
+        set(CMAKE_CXX_FLAGS_RELEASE "${CMAKE_CXX_FLAGS_RELEASE} -O3 -march=native -flto")
+        set(CMAKE_EXE_LINKER_FLAGS_RELEASE "${CMAKE_EXE_LINKER_FLAGS_RELEASE} -s")
+        set(CMAKE_SHARED_LINKER_FLAGS_RELEASE "${CMAKE_SHARED_LINKER_FLAGS_RELEASE} -s")
+    endif()
+endif()
+
+# Build options
+option(BUILD_SHARED_LIBS "Build libvideo2x as a shared library" ON)
+option(BUILD_VIDEO2X "Build the video2x executable" ON)
+option(USE_SYSTEM_NCNN "Use system ncnn library" ON)
+
+set(ALL_INCLUDE_DIRS)
+set(ALL_LIBRARIES)
+
+if(WIN32)
+    # Define base paths for FFmpeg and ncnn
+    set(FFMPEG_BASE_PATH ${PROJECT_SOURCE_DIR}/third_party/ffmpeg-shared)
+    set(NCNN_BASE_PATH ${PROJECT_SOURCE_DIR}/third_party/ncnn-shared/x64)
+
+    # FFmpeg
+    list(APPEND ALL_LIBRARIES
+        ${FFMPEG_BASE_PATH}/lib/avcodec.lib
+        ${FFMPEG_BASE_PATH}/lib/avdevice.lib
+        ${FFMPEG_BASE_PATH}/lib/avfilter.lib
+        ${FFMPEG_BASE_PATH}/lib/avformat.lib
+        ${FFMPEG_BASE_PATH}/lib/avutil.lib
+        ${FFMPEG_BASE_PATH}/lib/swscale.lib
+    )
+    list(APPEND ALL_INCLUDE_DIRS ${FFMPEG_BASE_PATH}/include)
+
+    # ncnn
+    # TODO: Figure out why this file is not being copied to the install directory
+    set(SPIRV_BUILD_PATH ${CMAKE_BINARY_DIR}/realesrgan-prefix/src/realesrgan-build/ncnn/glslang/SPIRV)
+    if (CMAKE_BUILD_TYPE STREQUAL "Release")
+        set(SPIRV_LIB ${SPIRV_BUILD_PATH}/Release/SPIRV.lib)
+    else()
+        set(SPIRV_LIB ${SPIRV_BUILD_PATH}/Debug/SPIRVd.lib)
+    endif()
+
+    list(APPEND ALL_LIBRARIES
+        ${NCNN_BASE_PATH}/lib/ncnn.lib
+        ${SPIRV_LIB}
+    )
+    list(APPEND ALL_INCLUDE_DIRS ${NCNN_BASE_PATH}/include/ncnn)
+else()
+    # Find the required packages using pkg-config
+    find_package(PkgConfig REQUIRED)
+    set(REQUIRED_PKGS
+        libavcodec
+        libavdevice
+        libavfilter
+        libavformat
+        libavutil
+        libswscale
+    )
+
+    # Loop through each package to find and collect include dirs and libraries
+    foreach(PKG ${REQUIRED_PKGS})
+        pkg_check_modules(${PKG} REQUIRED ${PKG})
+        list(APPEND ALL_INCLUDE_DIRS ${${PKG}_INCLUDE_DIRS})
+        list(APPEND ALL_LIBRARIES ${${PKG}_LIBRARIES})
+    endforeach()
+endif()
+
+# Remove duplicate entries
+list(REMOVE_DUPLICATES ALL_INCLUDE_DIRS)
+list(REMOVE_DUPLICATES ALL_LIBRARIES)
+
+# Find ncnn package
+if(USE_SYSTEM_NCNN)
+    find_package(ncnn REQUIRED)
+else()
+    option(NCNN_INSTALL_SDK "" OFF)
+    option(NCNN_PIXEL_ROTATE "" OFF)
+    option(NCNN_VULKAN "" ON)
+    option(NCNN_VULKAN_ONLINE_SPIRV "" ON)
+    option(NCNN_BUILD_BENCHMARK "" OFF)
+    option(NCNN_BUILD_TESTS "" OFF)
+    option(NCNN_BUILD_TOOLS "" OFF)
+    option(NCNN_BUILD_EXAMPLES "" OFF)
+    option(NCNN_DISABLE_RTTI "" ON)
+    option(NCNN_DISABLE_EXCEPTION "" ON)
+    option(NCNN_BUILD_SHARED_LIBS "" OFF)
+    option(SKIP_GLSLANG_INSTALL "" ON)
+
+    option(WITH_LAYER_absval "" OFF)
+    option(WITH_LAYER_argmax "" OFF)
+    option(WITH_LAYER_batchnorm "" OFF)
+    option(WITH_LAYER_bias "" OFF)
+    option(WITH_LAYER_bnll "" OFF)
+    option(WITH_LAYER_concat "" ON)
+    option(WITH_LAYER_convolution "" ON)
+    option(WITH_LAYER_crop "" ON)
+    option(WITH_LAYER_deconvolution "" OFF)
+    option(WITH_LAYER_dropout "" OFF)
+    option(WITH_LAYER_eltwise "" ON)
+    option(WITH_LAYER_elu "" OFF)
+    option(WITH_LAYER_embed "" OFF)
+    option(WITH_LAYER_exp "" OFF)
+    option(WITH_LAYER_flatten "" ON)
+    option(WITH_LAYER_innerproduct "" ON)
+    option(WITH_LAYER_input "" ON)
+    option(WITH_LAYER_log "" OFF)
+    option(WITH_LAYER_lrn "" OFF)
+    option(WITH_LAYER_memorydata "" OFF)
+    option(WITH_LAYER_mvn "" OFF)
+    option(WITH_LAYER_pooling "" OFF)
+    option(WITH_LAYER_power "" OFF)
+    option(WITH_LAYER_prelu "" ON)
+    option(WITH_LAYER_proposal "" OFF)
+    option(WITH_LAYER_reduction "" OFF)
+    option(WITH_LAYER_relu "" ON)
+    option(WITH_LAYER_reshape "" OFF)
+    option(WITH_LAYER_roipooling "" OFF)
+    option(WITH_LAYER_scale "" OFF)
+    option(WITH_LAYER_sigmoid "" OFF)
+    option(WITH_LAYER_slice "" OFF)
+    option(WITH_LAYER_softmax "" OFF)
+    option(WITH_LAYER_split "" ON)
+    option(WITH_LAYER_spp "" OFF)
+    option(WITH_LAYER_tanh "" OFF)
+    option(WITH_LAYER_threshold "" OFF)
+    option(WITH_LAYER_tile "" OFF)
+    option(WITH_LAYER_rnn "" OFF)
+    option(WITH_LAYER_lstm "" OFF)
+    option(WITH_LAYER_binaryop "" ON)
+    option(WITH_LAYER_unaryop "" OFF)
+    option(WITH_LAYER_convolutiondepthwise "" OFF)
+    option(WITH_LAYER_padding "" ON)
+    option(WITH_LAYER_squeeze "" OFF)
+    option(WITH_LAYER_expanddims "" OFF)
+    option(WITH_LAYER_normalize "" OFF)
+    option(WITH_LAYER_permute "" OFF)
+    option(WITH_LAYER_priorbox "" OFF)
+    option(WITH_LAYER_detectionoutput "" OFF)
+    option(WITH_LAYER_interp "" ON)
+    option(WITH_LAYER_deconvolutiondepthwise "" OFF)
+    option(WITH_LAYER_shufflechannel "" OFF)
+    option(WITH_LAYER_instancenorm "" OFF)
+    option(WITH_LAYER_clip "" OFF)
+    option(WITH_LAYER_reorg "" OFF)
+    option(WITH_LAYER_yolodetectionoutput "" OFF)
+    option(WITH_LAYER_quantize "" OFF)
+    option(WITH_LAYER_dequantize "" OFF)
+    option(WITH_LAYER_yolov3detectionoutput "" OFF)
+    option(WITH_LAYER_psroipooling "" OFF)
+    option(WITH_LAYER_roialign "" OFF)
+    option(WITH_LAYER_packing "" ON)
+    option(WITH_LAYER_requantize "" OFF)
+    option(WITH_LAYER_cast "" ON)
+    option(WITH_LAYER_hardsigmoid "" OFF)
+    option(WITH_LAYER_selu "" OFF)
+    option(WITH_LAYER_hardswish "" OFF)
+    option(WITH_LAYER_noop "" OFF)
+    option(WITH_LAYER_pixelshuffle "" ON)
+    option(WITH_LAYER_deepcopy "" OFF)
+    option(WITH_LAYER_mish "" OFF)
+    option(WITH_LAYER_statisticspooling "" OFF)
+    option(WITH_LAYER_swish "" OFF)
+    option(WITH_LAYER_gemm "" OFF)
+    option(WITH_LAYER_groupnorm "" OFF)
+    option(WITH_LAYER_layernorm "" OFF)
+    option(WITH_LAYER_softplus "" OFF)
+
+    add_subdirectory(third_party/ncnn)
+endif()
+
+# Include ExternalProject module
+include(ExternalProject)
+
+# Add libreal-esrgan-ncnn-vulkan as an external project
+ExternalProject_Add(
+    realesrgan
+    SOURCE_DIR ${PROJECT_SOURCE_DIR}/third_party/libreal_esrgan_ncnn_vulkan/src
+    CMAKE_ARGS
+        -DCMAKE_BUILD_TYPE=${CMAKE_BUILD_TYPE}
+        -DCMAKE_INSTALL_PREFIX=${CMAKE_BINARY_DIR}/realesrgan_install
+        -DCMAKE_POSITION_INDEPENDENT_CODE=ON
+        -DUSE_SYSTEM_NCNN=${USE_SYSTEM_NCNN}
+    BUILD_ALWAYS ON
+    INSTALL_COMMAND ${CMAKE_COMMAND} --build . --target install --config ${CMAKE_BUILD_TYPE}
+)
+
+# Add all source files for libvideo2x
+file(GLOB LIBVIDEO2X_SOURCES src/*.cpp)
+
+# Create the shared library 'libvideo2x'
+add_library(libvideo2x ${LIBVIDEO2X_SOURCES})
+target_compile_definitions(libvideo2x PRIVATE LIBVIDEO2X_EXPORTS)
+if(WIN32)
+    set_target_properties(libvideo2x PROPERTIES OUTPUT_NAME libvideo2x)
+else()
+    set_target_properties(libvideo2x PROPERTIES OUTPUT_NAME video2x)
+endif()
+
+# Ensure libvideo2x depends on realesrgan being built and installed
+add_dependencies(libvideo2x realesrgan)
+
+# Include directories for the shared library
+target_include_directories(libvideo2x PRIVATE
+    ${ALL_INCLUDE_DIRS}
+    ${PROJECT_SOURCE_DIR}/include
+    ${PROJECT_SOURCE_DIR}/third_party/libreal_esrgan_ncnn_vulkan/src
+)
+
+# Compile options for the shared library
+target_compile_options(libvideo2x PRIVATE
+    -Wall
+    -fPIC
+    $<$<CONFIG:Release>:-Ofast>
+    $<$<CONFIG:Debug>:-g -DDEBUG>
+)
+
+# Define the path to the built libresrgan-ncnn-vulkan library
+if(WIN32)
+    set(REALESRGAN_LIB ${CMAKE_BINARY_DIR}/realesrgan_install/lib/realesrgan-ncnn-vulkan.lib)
+else()
+    set(REALESRGAN_LIB ${CMAKE_BINARY_DIR}/realesrgan_install/lib/librealesrgan-ncnn-vulkan.so)
+endif()
+
+# Link the shared library with the dependencies
+target_link_libraries(libvideo2x PRIVATE ${ALL_LIBRARIES} ${REALESRGAN_LIB})
+
+if(NOT WIN32)
+    if (USE_SYSTEM_NCNN)
+        target_link_libraries(libvideo2x PUBLIC ncnn)
+    else()
+        target_link_libraries(libvideo2x PRIVATE ncnn)
+    endif()
+endif()
+
+# Create the executable 'video2x'
+if (BUILD_VIDEO2X)
+    add_executable(video2x src/video2x.c src/getopt.c)
+    set_target_properties(video2x PROPERTIES OUTPUT_NAME video2x)
+
+    # Include directories for the executable
+    target_include_directories(video2x PRIVATE
+        ${ALL_INCLUDE_DIRS}
+        ${PROJECT_SOURCE_DIR}/include
+    )
+
+    # Compile options for the executable
+    target_compile_options(video2x PRIVATE
+        -Wall
+        $<$<CONFIG:Debug>:-g -DDEBUG>
+    )
+
+    # Link the executable with the shared library
+    target_link_libraries(video2x PRIVATE ${ALL_LIBRARIES} libvideo2x)
+endif()
+
+# Define the default installation directories
+if(WIN32)
+    set(BIN_DESTINATION_DEFAULT ".")
+    set(INCLUDE_DESTINATION_DEFAULT "include")
+    set(LIB_DESTINATION_DEFAULT ".")
+    set(MODEL_DESTINATION_DEFAULT ".")
+else()
+    set(BIN_DESTINATION_DEFAULT "bin")
+    set(INCLUDE_DESTINATION_DEFAULT "include")
+    set(LIB_DESTINATION_DEFAULT "lib")
+    set(MODEL_DESTINATION_DEFAULT "share/video2x")
+endif()
+
+# Set the installation directories
+set(INSTALL_BIN_DESTINATION ${BIN_DESTINATION_DEFAULT} CACHE STRING "")
+set(INSTALL_INCLUDE_DESTINATION ${INCLUDE_DESTINATION_DEFAULT} CACHE STRING "")
+set(INSTALL_LIB_DESTINATION ${LIB_DESTINATION_DEFAULT} CACHE STRING "")
+set(INSTALL_MODEL_DESTINATION ${MODEL_DESTINATION_DEFAULT} CACHE STRING "")
+
+# Common installation rules for libvideo2x and models
+install(TARGETS libvideo2x
+    LIBRARY DESTINATION ${INSTALL_LIB_DESTINATION}
+        PERMISSIONS OWNER_READ OWNER_WRITE OWNER_EXECUTE
+                    GROUP_READ GROUP_EXECUTE
+                    WORLD_READ WORLD_EXECUTE
+    ARCHIVE DESTINATION ${INSTALL_LIB_DESTINATION}
+    RUNTIME DESTINATION ${INSTALL_BIN_DESTINATION}
+)
+
+# Install model files
+install(DIRECTORY ${CMAKE_SOURCE_DIR}/models DESTINATION ${INSTALL_MODEL_DESTINATION})
+
+# Install the executable if BUILD_VIDEO2X is enabled
+if(BUILD_VIDEO2X)
+    install(TARGETS video2x RUNTIME DESTINATION ${INSTALL_BIN_DESTINATION})
+endif()
+
+# Install the header file
+install(FILES ${PROJECT_SOURCE_DIR}/include/libvideo2x.h DESTINATION ${INSTALL_INCLUDE_DESTINATION})
+
+# Platform-specific installation rules
+if(WIN32)
+    # Install Windows-specific dependencies
+    install(FILES ${CMAKE_BINARY_DIR}/realesrgan_install/bin/realesrgan-ncnn-vulkan.dll
+        DESTINATION ${INSTALL_BIN_DESTINATION}
+        PERMISSIONS OWNER_READ OWNER_WRITE OWNER_EXECUTE
+                    GROUP_READ GROUP_EXECUTE
+                    WORLD_READ WORLD_EXECUTE
+    )
+    install(FILES ${NCNN_BASE_PATH}/bin/ncnn.dll DESTINATION ${INSTALL_BIN_DESTINATION}
+        PERMISSIONS OWNER_READ OWNER_WRITE OWNER_EXECUTE
+                    GROUP_READ GROUP_EXECUTE
+                    WORLD_READ WORLD_EXECUTE
+    )
+else()
+    # Install Unix-specific dependencies
+    install(FILES ${REALESRGAN_LIB} DESTINATION ${INSTALL_LIB_DESTINATION}
+        PERMISSIONS OWNER_READ OWNER_WRITE OWNER_EXECUTE
+                    GROUP_READ GROUP_EXECUTE
+                    WORLD_READ WORLD_EXECUTE
+    )
+endif()

CNAME

@@ -1 +0,0 @@
-video2x.org

Dockerfile

@@ -0,0 +1,45 @@
+# Name: Video2X Dockerfile
+# Creator: K4YT3X
+# Date Created: February 3, 2022
+# Last Modified: October 7, 2024
+
+# stage 1: build the python components into wheels
+FROM docker.io/archlinux:latest AS builder
+
+# Install dependencies and create a non-root user
+RUN pacman -Syy --noconfirm \
+        base-devel ffmpeg ncnn git cmake make clang pkgconf vulkan-headers openmp sudo \
+        nvidia-utils vulkan-radeon vulkan-intel vulkan-swrast \
+    && useradd -m builder \
+    && echo 'builder ALL=(ALL) NOPASSWD: ALL' > /etc/sudoers.d/builder
+
+# Switch to the non-root user and copy the source code
+USER builder
+COPY --chown=builder:builder . /video2x
+WORKDIR /video2x
+
+# Build the package
+RUN makepkg -s --noconfirm \
+    && find /video2x -maxdepth 1 -name '*.pkg.tar.zst' | head -n 1 | \
+        xargs -I {} cp {} /tmp/video2x.pkg.tar.zst
+
+# stage 2: install wheels into the final image
+FROM docker.io/archlinux:latest
+LABEL maintainer="K4YT3X <i@k4yt3x.com>" \
+      org.opencontainers.image.source="https://github.com/k4yt3x/video2x" \
+      org.opencontainers.image.description="A lossless video super resolution framework"
+
+ENV VK_ICD_FILENAMES=/usr/share/vulkan/icd.d/nvidia_icd.json\
+:/usr/share/vulkan/icd.d/radeon_icd.x86_64.json\
+:/usr/share/vulkan/icd.d/intel_icd.x86_64.json
+
+COPY --from=builder /tmp/video2x.pkg.tar.zst /video2x.pkg.tar.zst
+COPY . /video2x
+WORKDIR /video2x
+RUN pacman -Sy --noconfirm ffmpeg ncnn \
+        nvidia-utils vulkan-radeon vulkan-intel vulkan-swrast \
+    && pacman -U --noconfirm /video2x.pkg.tar.zst \
+    && rm -rf /video2x.pkg.tar.zst /var/cache/pacman/pkg/*
+
+WORKDIR /host
+ENTRYPOINT ["/usr/bin/video2x"]

LICENSE

@@ -1,23 +1,21 @@
-                    GNU GENERAL PUBLIC LICENSE
-                       Version 3, 29 June 2007
+                    GNU AFFERO GENERAL PUBLIC LICENSE
+                       Version 3, 19 November 2007
 
- Copyright (C) 2007 Free Software Foundation, Inc. <http://fsf.org/>
+ Copyright (C) 2007 Free Software Foundation, Inc. <https://fsf.org/>
  Everyone is permitted to copy and distribute verbatim copies
  of this license document, but changing it is not allowed.
 
                             Preamble
 
-  The GNU General Public License is a free, copyleft license for
-software and other kinds of works.
+  The GNU Affero General Public License is a free, copyleft license for
+software and other kinds of works, specifically designed to ensure
+cooperation with the community in the case of network server software.
 
   The licenses for most software and other practical works are designed
 to take away your freedom to share and change the works.  By contrast,
-the GNU General Public License is intended to guarantee your freedom to
+our General Public Licenses are intended to guarantee your freedom to
 share and change all versions of a program--to make sure it remains free
-software for all its users.  We, the Free Software Foundation, use the
-GNU General Public License for most of our software; it applies also to
-any other work released this way by its authors.  You can apply it to
-your programs, too.
+software for all its users.
 
   When we speak of free software, we are referring to freedom, not
 price.  Our General Public Licenses are designed to make sure that you
@@ -26,44 +24,34 @@ them if you wish), that you receive source code or can get it if you
 want it, that you can change the software or use pieces of it in new
 free programs, and that you know you can do these things.
 
-  To protect your rights, we need to prevent others from denying you
-these rights or asking you to surrender the rights.  Therefore, you have
-certain responsibilities if you distribute copies of the software, or if
-you modify it: responsibilities to respect the freedom of others.
+  Developers that use our General Public Licenses protect your rights
+with two steps: (1) assert copyright on the software, and (2) offer
+you this License which gives you legal permission to copy, distribute
+and/or modify the software.
 
-  For example, if you distribute copies of such a program, whether
-gratis or for a fee, you must pass on to the recipients the same
-freedoms that you received.  You must make sure that they, too, receive
-or can get the source code.  And you must show them these terms so they
-know their rights.
+  A secondary benefit of defending all users' freedom is that
+improvements made in alternate versions of the program, if they
+receive widespread use, become available for other developers to
+incorporate.  Many developers of free software are heartened and
+encouraged by the resulting cooperation.  However, in the case of
+software used on network servers, this result may fail to come about.
+The GNU General Public License permits making a modified version and
+letting the public access it on a server without ever releasing its
+source code to the public.
 
-  Developers that use the GNU GPL protect your rights with two steps:
-(1) assert copyright on the software, and (2) offer you this License
-giving you legal permission to copy, distribute and/or modify it.
+  The GNU Affero General Public License is designed specifically to
+ensure that, in such cases, the modified source code becomes available
+to the community.  It requires the operator of a network server to
+provide the source code of the modified version running there to the
+users of that server.  Therefore, public use of a modified version, on
+a publicly accessible server, gives the public access to the source
+code of the modified version.
 
-  For the developers' and authors' protection, the GPL clearly explains
-that there is no warranty for this free software.  For both users' and
-authors' sake, the GPL requires that modified versions be marked as
-changed, so that their problems will not be attributed erroneously to
-authors of previous versions.
-
-  Some devices are designed to deny users access to install or run
-modified versions of the software inside them, although the manufacturer
-can do so.  This is fundamentally incompatible with the aim of
-protecting users' freedom to change the software.  The systematic
-pattern of such abuse occurs in the area of products for individuals to
-use, which is precisely where it is most unacceptable.  Therefore, we
-have designed this version of the GPL to prohibit the practice for those
-products.  If such problems arise substantially in other domains, we
-stand ready to extend this provision to those domains in future versions
-of the GPL, as needed to protect the freedom of users.
-
-  Finally, every program is threatened constantly by software patents.
-States should not allow patents to restrict development and use of
-software on general-purpose computers, but in those that do, we wish to
-avoid the special danger that patents applied to a free program could
-make it effectively proprietary.  To prevent this, the GPL assures that
-patents cannot be used to render the program non-free.
+  An older license, called the Affero General Public License and
+published by Affero, was designed to accomplish similar goals.  This is
+a different license, not a version of the Affero GPL, but Affero has
+released a new version of the Affero GPL which permits relicensing under
+this license.
 
   The precise terms and conditions for copying, distribution and
 modification follow.
@@ -72,7 +60,7 @@ modification follow.
 
   0. Definitions.
 
-  "This License" refers to version 3 of the GNU General Public License.
+  "This License" refers to version 3 of the GNU Affero General Public License.
 
   "Copyright" also means copyright-like laws that apply to other kinds of
 works, such as semiconductor masks.
@@ -549,35 +537,45 @@ to collect a royalty for further conveying from those to whom you convey
 the Program, the only way you could satisfy both those terms and this
 License would be to refrain entirely from conveying the Program.
 
-  13. Use with the GNU Affero General Public License.
+  13. Remote Network Interaction; Use with the GNU General Public License.
+
+  Notwithstanding any other provision of this License, if you modify the
+Program, your modified version must prominently offer all users
+interacting with it remotely through a computer network (if your version
+supports such interaction) an opportunity to receive the Corresponding
+Source of your version by providing access to the Corresponding Source
+from a network server at no charge, through some standard or customary
+means of facilitating copying of software.  This Corresponding Source
+shall include the Corresponding Source for any work covered by version 3
+of the GNU General Public License that is incorporated pursuant to the
+following paragraph.
 
   Notwithstanding any other provision of this License, you have
 permission to link or combine any covered work with a work licensed
-under version 3 of the GNU Affero General Public License into a single
+under version 3 of the GNU General Public License into a single
 combined work, and to convey the resulting work.  The terms of this
 License will continue to apply to the part which is the covered work,
-but the special requirements of the GNU Affero General Public License,
-section 13, concerning interaction through a network will apply to the
-combination as such.
+but the work with which it is combined will remain governed by version
+3 of the GNU General Public License.
 
   14. Revised Versions of this License.
 
   The Free Software Foundation may publish revised and/or new versions of
-the GNU General Public License from time to time.  Such new versions will
-be similar in spirit to the present version, but may differ in detail to
+the GNU Affero General Public License from time to time.  Such new versions
+will be similar in spirit to the present version, but may differ in detail to
 address new problems or concerns.
 
   Each version is given a distinguishing version number.  If the
-Program specifies that a certain numbered version of the GNU General
+Program specifies that a certain numbered version of the GNU Affero General
 Public License "or any later version" applies to it, you have the
 option of following the terms and conditions either of that numbered
 version or of any later version published by the Free Software
 Foundation.  If the Program does not specify a version number of the
-GNU General Public License, you may choose any version ever published
+GNU Affero General Public License, you may choose any version ever published
 by the Free Software Foundation.
 
   If the Program specifies that a proxy can decide which future
-versions of the GNU General Public License can be used, that proxy's
+versions of the GNU Affero General Public License can be used, that proxy's
 public statement of acceptance of a version permanently authorizes you
 to choose that version for the Program.
 
@@ -635,40 +633,29 @@ the "copyright" line and a pointer to where the full notice is found.
     Copyright (C) <year>  <name of author>
 
     This program is free software: you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
+    it under the terms of the GNU Affero General Public License as published by
     the Free Software Foundation, either version 3 of the License, or
     (at your option) any later version.
 
     This program is distributed in the hope that it will be useful,
     but WITHOUT ANY WARRANTY; without even the implied warranty of
     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
+    GNU Affero General Public License for more details.
 
-    You should have received a copy of the GNU General Public License
-    along with this program.  If not, see <http://www.gnu.org/licenses/>.
+    You should have received a copy of the GNU Affero General Public License
+    along with this program.  If not, see <https://www.gnu.org/licenses/>.
 
 Also add information on how to contact you by electronic and paper mail.
 
-  If the program does terminal interaction, make it output a short
-notice like this when it starts in an interactive mode:
-
-    <program>  Copyright (C) <year>  <name of author>
-    This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
-    This is free software, and you are welcome to redistribute it
-    under certain conditions; type `show c' for details.
-
-The hypothetical commands `show w' and `show c' should show the appropriate
-parts of the General Public License.  Of course, your program's commands
-might be different; for a GUI interface, you would use an "about box".
+  If your software can interact with users remotely through a computer
+network, you should also make sure that it provides a way for users to
+get its source.  For example, if your program is a web application, its
+interface could display a "Source" link that leads users to an archive
+of the code.  There are many ways you could offer source, and different
+solutions will be better for different programs; see section 13 for the
+specific requirements.
 
   You should also get your employer (if you work as a programmer) or school,
 if any, to sign a "copyright disclaimer" for the program, if necessary.
-For more information on this, and how to apply and follow the GNU GPL, see
-<http://www.gnu.org/licenses/>.
-
-  The GNU General Public License does not permit incorporating your program
-into proprietary programs.  If your program is a subroutine library, you
-may consider it more useful to permit linking proprietary applications with
-the library.  If this is what you want to do, use the GNU Lesser General
-Public License instead of this License.  But first, please read
-<http://www.gnu.org/philosophy/why-not-lgpl.html>.
+For more information on this, and how to apply and follow the GNU AGPL, see
+<https://www.gnu.org/licenses/>.

Makefile

@@ -0,0 +1,71 @@
+.PHONY: build static debug windows test-realesrgan test-libplacebo leakcheck clean
+
+BINDIR=build
+CC=clang
+CXX=clang++
+
+build:
+	cmake -S . -B $(BINDIR) \
+		-DCMAKE_EXPORT_COMPILE_COMMANDS=ON \
+		-DCMAKE_C_COMPILER=$(CC) \
+		-DCMAKE_CXX_COMPILER=$(CXX) \
+		-DCMAKE_BUILD_TYPE=Release
+	cmake --build $(BINDIR) --config Release --parallel
+	cp $(BINDIR)/compile_commands.json .
+
+static:
+	cmake -S . -B $(BINDIR) \
+		-DCMAKE_EXPORT_COMPILE_COMMANDS=ON \
+		-DCMAKE_C_COMPILER=$(CC) \
+		-DCMAKE_CXX_COMPILER=$(CXX) \
+		-DCMAKE_BUILD_TYPE=Release \
+		-DBUILD_SHARED_LIBS=OFF \
+		-DUSE_SYSTEM_NCNN=OFF
+	cmake --build $(BINDIR) --config Release --parallel
+	cp $(BINDIR)/compile_commands.json .
+
+debug:
+	cmake -S . -B $(BINDIR) \
+		-DCMAKE_EXPORT_COMPILE_COMMANDS=ON \
+		-DCMAKE_C_COMPILER=$(CC) \
+		-DCMAKE_CXX_COMPILER=$(CXX) \
+		-DCMAKE_BUILD_TYPE=Debug
+	cmake --build $(BINDIR) --config Debug --parallel
+	cp $(BINDIR)/compile_commands.json .
+
+test-realesrgan:
+	LD_LIBRARY_PATH=$(BINDIR) $(BINDIR)/video2x -i data/standard-test.mp4 -o data/output.mp4 \
+		-f realesrgan -r 4 --model realesr-animevideov3
+
+test-libplacebo:
+	LD_LIBRARY_PATH=$(BINDIR) $(BINDIR)/video2x -i data/standard-test.mp4 -o data/output.mp4 \
+		-f libplacebo -w 1920 -h 1080 -s anime4k-mode-a
+
+leakcheck-realesrgan:
+	LD_LIBRARY_PATH=$(BINDIR) valgrind \
+		--tool=memcheck \
+		--leak-check=full \
+		--show-leak-kinds=all \
+		--track-origins=yes \
+		--show-reachable=yes \
+		--verbose --log-file="valgrind.log" \
+		$(BINDIR)/video2x \
+		-i data/standard-test.mp4 -o data/output.mp4 \
+		-f realesrgan -r 2 --model realesr-animevideov3 \
+		-p veryfast -b 1000000 -q 30
+
+leakcheck-libplacebo:
+	LD_LIBRARY_PATH=$(BINDIR) valgrind \
+		--tool=memcheck \
+		--leak-check=full \
+		--show-leak-kinds=all \
+		--track-origins=yes \
+		--show-reachable=yes \
+		--verbose --log-file="valgrind.log" \
+		$(BINDIR)/video2x \
+		-i data/standard-test.mp4 -o data/output.mp4 \
+		-f libplacebo -w 1920 -h 1080 -s anime4k-mode-a \
+		-p veryfast -b 1000000 -q 30
+
+clean:
+	rm -rf $(BINDIR)

NOTICE

@@ -0,0 +1,11 @@
+Video2X
+Copyright (C) 2018-2024 K4YT3X and contributors.
+
+This product depends on FFmpeg, which is available under the GNU Lesser General Public License 2.1.
+The source code can be found at https://github.com/FFmpeg/FFmpeg.
+
+This product depends on Real-ESRGAN ncnn Vulkan, which is available under the MIT License.
+The source code can be found at https://github.com/xinntao/Real-ESRGAN-ncnn-vulkan.
+
+This product depends on ncnn, which is available under the BSD 3-Clause License.
+The source code can be found at https://github.com/Tencent/ncnn.

PKGBUILD

@@ -0,0 +1,27 @@
+pkgname=video2x
+pkgver=r862.f590ead
+pkgrel=1
+pkgdesc="A lossless video super resolution framework"
+arch=('x86_64')
+url="https://github.com/k4yt3x/video2x"
+license=('AGPL3')
+depends=('ffmpeg' 'ncnn' 'vulkan-driver')
+makedepends=('git' 'cmake' 'make' 'clang' 'pkgconf' 'vulkan-headers' 'openmp')
+
+pkgver() {
+    printf "r%s.%s" "$(git rev-list --count HEAD)" "$(git rev-parse --short HEAD)"
+}
+
+prepare() {
+    git submodule update --init --recursive
+}
+
+build() {
+    cmake -B build -S .. -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=/usr
+    cmake --build build --config Release --parallel
+}
+
+package() {
+    DESTDIR="$pkgdir" cmake --install build
+}
+

README.md

@@ -1,272 +1,121 @@
 <p align="center">
-   <img src="https://user-images.githubusercontent.com/21986859/81489504-c7d1f780-9265-11ea-86c8-cc0316e2082d.png"/>
+   <img src="https://github.com/user-attachments/assets/0c6c5d31-7fd5-4e50-b41e-91a58aae995c"/>
+   </br>
+   <img src="https://img.shields.io/github/v/release/k4yt3x/video2x?style=flat-square"/>
+   <img src="https://img.shields.io/github/actions/workflow/status/k4yt3x/video2x/build.yml?label=Build&style=flat-square"/>
+   <img src="https://img.shields.io/github/downloads/k4yt3x/video2x/total?style=flat-square"/>
+   <img src="https://img.shields.io/github/license/k4yt3x/video2x?style=flat-square"/>
+   <img src="https://img.shields.io/badge/dynamic/json?color=%23e85b46&label=Patreon&query=data.attributes.patron_count&suffix=%20patrons&url=https%3A%2F%2Fwww.patreon.com%2Fapi%2Fcampaigns%2F4507807&style=flat-square"/>
 </p>
 
-![Master Branch Version](https://img.shields.io/badge/master-v4.0.0-9cf?style=flat-square)
-![GitHub release (latest by date)](https://img.shields.io/github/v/release/k4yt3x/video2x?style=flat-square)
-![GitHub All Releases](https://img.shields.io/github/downloads/k4yt3x/video2x/total?style=flat-square)
-![GitHub](https://img.shields.io/github/license/k4yt3x/video2x?style=flat-square)
-![Platforms](https://img.shields.io/badge/Platforms-Windows%20%7C%20Linux%20%7C%20macOS-blue?style=flat-square)
-<img alt="Become a Patron!"
-src="https://c5.patreon.com/external/logo/become_a_patron_button@2x.png"
-href="https://www.patreon.com/bePatron?u=34970782"
-height=20 />
+## 🌟 Version 6.0.0 Preview
 
-<!--# Video2X Lossless Video Enlarger-->
+Version 6.0.0 is a complete rewrite of this project in C/C++. It:
 
-### Official Discussion Group (Telegram): https://t.me/video2x
+- actually works, with less pain;
+- is blazing fast, thanks to the redesigned efficient pipeline and the speed of C/C++;
+- is cross-platform, available right now for both Windows and Linux;
+- supports Anime4K v4 and all other custom MPV-compatible GLSL shaders;
+- supports RealESRGAN (all three models) via ncnn and Vulkan;
+- requires 0 disk space for processing the video, just space for storing the final output; and
+- exports a standard C function that can be easily integrated in your own projects!
 
-## [Download Builds](https://github.com/k4yt3x/video2x/releases) (Windows)
+These are available for download now:
 
-You can go to the [releases page](https://github.com/k4yt3x/video2x/releases) to download the latest builds of Video2X. The exe files will require no Python or Python module installation. There are two editions available.
+- 6.0.0 beta CLI preview builds for Windows and Linux are on the [releases page](https://github.com/k4yt3x/video2x/releases).
+  - You will need to install the dependencies and set `LD_LIBRARY_PATH` for the Linux build to work. Refer to the [PKGBUILD](PKGBUILD) file to see what needs to be installed.
+  - Alternatively, you can build it from source. Take a look at the [Makefile](Makefile).
+- 6.0.0 beta AUR package for Arch Linux (`video2x-git`).
+- 6.0.0 beta [container image](https://github.com/k4yt3x/video2x/pkgs/container/video2x).
+- A GUI for 6.0.0 is in the making. More information about it will be available soon.
 
-- **`Full`**: full package comes pre-configured with **all** dependencies like `FFmpeg` and `waifu2x-caffe`.
-- **`Light`**: ligt package comes with only Video2X binaries and a template configuration file. The user will either have to run the setup script or install and configure dependencies themselves.
+There is still much to be done and optimize. Stay tuned for more updates. As for why the 5.0.0 branch was abandoned, here are some of the reasons:
 
-Go to the [Quick Start](#quick-start) section for usages.
+- Wrapped C++ libraries for Python is too painful to build for cross-platform distribution.
+- Some wrapped C++ libraires exhibited unexpected behaviors.
+- Running FFmpeg via commands and piping data through stdin/stdout are inefficient.
+- C/C++ native binaries are much smaller and much more efficient.
+
+## [💬 Telegram Discussion Group](https://t.me/video2x)
+
+Join our Telegram discussion group to ask any questions you have about Video2X, chat directly with the developers, or discuss about super resolution technologies and the future of Video2X in general.
+
+## [🪟 Download Windows Releases](https://github.com/k4yt3x/video2x/releases/tag/4.8.1)
+
+The latest Windows update is built based on version 4.8.1. GUI is not available for 6.0.0 yet, but is already under development. Go to the [GUI](https://github.com/k4yt3x/video2x/wiki/GUI) page to see the basic usages of the GUI. Try the [mirror](https://files.k4yt3x.com/Projects/Video2X/latest) if you can't download releases directly from GitHub.
+
+## [📔 Google Colab](https://colab.research.google.com/drive/1gWEwcA9y57EsxwOjmLNmNMXPsafw0kGo)
+
+You can use Video2X on [Google Colab](https://colab.research.google.com/) **for free** if you don't have a powerful GPU of your own. You can borrow a powerful GPU (Tesla K80, T4, P4, or P100) on Google's server for free for a maximum of 12 hours per session. **Please use the free resource fairly** and do not create sessions back-to-back and run upscaling 24/7. This might result in you getting banned. You can get [Colab Pro/Pro+](https://colab.research.google.com/signup/pricing) if you'd like to use better GPUs and get longer runtimes. Usage instructions are embedded in the [Colab Notebook](https://colab.research.google.com/drive/1gWEwcA9y57EsxwOjmLNmNMXPsafw0kGo).
+
+## [📦 Container Image](https://github.com/k4yt3x/video2x/pkgs/container/video2x)
+
+Video2X container images are available on the GitHub Container Registry for easy deployment on Linux and macOS. If you already have Docker/Podman installed, only one command is needed to start upscaling a video. For more information on how to use Video2X's Docker image, please refer to the [documentations](https://github.com/K4YT3X/video2x/wiki/Container).
+
+## [📖 Documentations](https://github.com/k4yt3x/video2x/wiki)
+
+Video2X's documentations are hosted on this repository's [Wiki page](https://github.com/k4yt3x/video2x/wiki). It includes comprehensive explanations for how to use the [GUI](https://github.com/k4yt3x/video2x/wiki/GUI), the [CLI](https://github.com/k4yt3x/video2x/wiki/CLI), the [container image](https://github.com/K4YT3X/video2x/wiki/Container), the [library](https://github.com/k4yt3x/video2x/wiki/Library), and more. The Wiki is open to edits by the community, so you, yes you, can also correct errors or add new contents to the documentations.
 
 ## Introduction
 
-Video2X is a video upscaling software based on Waifu2X, Anime4K and SRMD written in Python 3. It upscales videos and restores details from low-resolution videos. Below is a side-by-side preview.
+Video2X is a video/GIF/image upscaling and frame interpolation software written in Python. It can use these following state-of-the-art algorithms to increase the resolution and frame rate of your video/GIF/image. More information about the algorithms that it supports can be found in [the documentations](https://github.com/k4yt3x/video2x/wiki/Algorithms).
 
-![Spirited Away Demo](https://user-images.githubusercontent.com/21986859/49412428-65083280-f73a-11e8-8237-bb34158a545e.png)
+### Video Upscaling
 
-*Upscale Comparison Demonstration*
+![Spirited Away Demo](https://user-images.githubusercontent.com/21986859/49412428-65083280-f73a-11e8-8237-bb34158a545e.png)\
+_Upscale demo: Spirited Away's movie trailer_
 
-**You can watch the whole demo video on YouTube: https://youtu.be/mGEfasQl2Zo**
+- **Spirited Away**: [YouTube](https://youtu.be/mGEfasQl2Zo) | [Bilibili](https://www.bilibili.com/video/BV1V5411471i/)
+  - 360P to 4K
+  - The [original video](https://www.youtube.com/watch?v=ByXuk9QqQkk)'s copyright belongs to 株式会社スタジオジブリ
+- **Bad Apple!!**: [YouTube](https://youtu.be/A81rW_FI3cw) | [Bilibili](https://www.bilibili.com/video/BV16K411K7ue)
+  - 384P to 4K 120FPS
+  - The [original video](https://www.nicovideo.jp/watch/sm8628149)'s copyright belongs to あにら
+- **The Pet Girl of Sakurasou**: [YouTube](https://youtu.be/M0vDI1HH2_Y) | [Bilibili](https://www.bilibili.com/video/BV14k4y167KP/)
+  - 240P to 1080P 60FPS
+  - The original video's copyright belongs to ASCII Media Works
 
-Clip is from trailer of animated movie "千と千尋の神隠し". Copyright belongs to "株式会社スタジオジブリ (STUDIO GHIBLI INC.)". Will delete immediately if use of clip is in violation of copyright.
+### Standard Test Clip
 
-## Demo Videos
+The following clip can be used to test if your setup works properly. This is also the standard clip used for running performance benchmarks.
 
-Below is a list of all the demo videos available.
+- [Standard Test Clip (240P)](https://files.k4yt3x.com/Resources/Videos/standard-test.mp4) 4.54 MiB
+- [waifu2x Upscaled Sample (1080P)](https://files.k4yt3x.com/Resources/Videos/standard-waifu2x.mp4) 4.54 MiB
+- [Ground Truth (1080P)](https://files.k4yt3x.com/Resources/Videos/standard-original.mp4) 22.2 MiB
 
-- **Spirited Away (360P to 4K)**
-  - Original name: 千と千尋の神隠し
-  - YouTube: https://youtu.be/mGEfasQl2Zo
-  - Bilibili: https://www.bilibili.com/video/BV1V5411471i/
-- **The Pet Girl of Sakurasou 240P to 1080P 60FPS**
-  - Original name: さくら荘のペットな彼女
-  - YouTube: https://youtu.be/M0vDI1HH2_Y
-  - Bilibili: https://www.bilibili.com/video/BV14k4y167KP/
-
-## Screenshots
-
-### Video2X GUI
-
-![GUI Preview](https://user-images.githubusercontent.com/21986859/81546668-3bf5c380-936a-11ea-9583-c969ea0d862b.png)
-
-*Video2X GUI Screenshot*
-
-### Video2X CLI
-
-![Video2X CLI Screenshot](https://user-images.githubusercontent.com/21986859/81039711-4fe88380-8e99-11ea-9846-175f72100a76.png)
-
-*Video2X CLI Screenshot*
-
----
-
-## Documentations
-
-### [Video2X Wiki](https://github.com/k4yt3x/video2x/wiki)
-
-You can find all detailed user-facing and developer-facing documentations in the [Video2X Wiki](https://github.com/k4yt3x/video2x/wiki). It covers everything from step-by-step instructions for beginners, to the code structure of this program for advanced users and developers. If this README page doesn't answer all your questions, the wiki page is where you should head to.
-
-### [Run From Source](https://github.com/k4yt3x/video2x/wiki/Run-From-Source)
-
-Instructions for how to run this program from source code.
-
-### [Step-By-Step Tutorial](https://github.com/k4yt3x/video2x/wiki/Step-By-Step-Tutorial)
-
-For those who want a detailed walk-through of how to use Video2X, you can head to the [Step-By-Step Tutorial](https://github.com/k4yt3x/video2x/wiki/Step-By-Step-Tutorial) wiki page. It includes almost every step you need to perform in order to enlarge your first video.
-
-### [Drivers](https://github.com/k4yt3x/video2x/wiki/Drivers)
-
-Go to the [Drivers](https://github.com/k4yt3x/video2x/wiki/Drivers) wiki page if you want to see a detailed description on the different types of drivers implemented by Video2X. This wiki page contains detailed difference between different drivers, and how to download and set each of them up for Video2X.
-
-### [Q&A](https://github.com/k4yt3x/video2x/wiki/Q&A)
-
-If you have any questions, first try visiting our [Q&A](https://github.com/k4yt3x/video2x/wiki/Q&A) page to see if your question is answered there. If not, open an issue and we will respond to your questions ASAP. Alternatively, you can also join our [Telegram discussion group](https://t.me/video2x) and ask your questions there.
-
----
-
-### Sample Videos
-
-If you can't find a video clip to begin with, or if you want to see a before-after comparison, we have prepared some sample clips for you. The quick start guide down below will also be based on the name of the sample clips.
-
-![sample_video](https://user-images.githubusercontent.com/21986859/52905766-d5512b00-3236-11e9-9aea-077636539679.png)
-
-*Sample Upscale Videos*
-
-- [Sample Video (240P) 4.54MB](https://files.k4yt3x.com/Resources/Videos/sample_input.mp4)
-- [Sample Video Upscaled (1080P) 4.54MB](https://files.k4yt3x.com/Resources/Videos/sample_output.mp4)
-- [Sample Video Original (1080P) 22.2MB](https://files.k4yt3x.com/Resources/Videos/sample_original.mp4)
-
-Clip is from anime "さくら荘のペットな彼女". Copyright belongs to "株式会社アニプレックス (Aniplex Inc.)". Will delete immediately if use of clip is in violation of copyright.
-
----
-
-## Quick Start
-
-### Prerequisites
-
-Before running Video2X, you'll need to ensure you have installed the drivers' external dependencies such as GPU drivers.
-
-- waifu2x-caffe
-  - GPU mode: Nvidia graphics card driver
-  - cuDNN mode: Nvidia CUDA and [cuDNN](https://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows)
-- Other Drivers
-  - GPU driver if you want to use GPU for processing
-
-### Running Video2X (GUI)
-
-The easiest way to run Video2X is to use the full build. Extract the full release zip file and you'll get these files.
-
-![Video2X Release Files](https://user-images.githubusercontent.com/21986859/81489846-28633380-926a-11ea-9e81-fb92f492e14c.png)
-
-Simply double click on video2x_gui.exe to launch the GUI.
-
-![Video2X GUI Main Tab](https://user-images.githubusercontent.com/21986859/81489858-4c267980-926a-11ea-9ab2-38ec738f2fb6.png)
-
-Then, drag the videos you wish to upscale into the window and select the appropriate output path.
-
-![drag-drop](https://user-images.githubusercontent.com/21986859/81489880-7bd58180-926a-11ea-85ae-b72d2f4f5e72.png)
-
-Tweak the settings if you want to, then hit the start button at the bottom and the upscale will start. Now you'll just have to wait for it to complete.
-
-![upscale-started](https://user-images.githubusercontent.com/21986859/81489924-ce16a280-926a-11ea-831c-6c66b950f957.png)
-
-### Running Video2X (CLI)
-
-#### Basic Upscale Example
-
-This example command below uses `waifu2x-caffe` to enlarge the video `sample-input.mp4` two double its original size.
-
-```shell
-python video2x.py -i sample-input.mp4 -o sample-output.mp4 -r 2 -d waifu2x_caffe
-```
-
-#### Advanced Upscale Example
-
-If you would like to tweak engine-specific settings, either specify the corresponding argument after `--`, or edit the corresponding field in the configuration file `video2x.yaml`. **Command line arguments will overwrite default values in the config file.**
-
-This example below adds enables TTA for `waifu2x-caffe`.
-
-```shell
-python video2x.py -i sample-input.mp4 -o sample-output.mp4 -r 2 -d waifu2x_caffe -- --tta 1
-```
-
-To see a help page for driver-specific settings, use `-d` to select the driver and append `-- --help` as demonstrated below. This will print all driver-specific settings and descriptions.
-
-```shell
-python video2x.py -d waifu2x_caffe -- --help
-```
-
----
-
-## Recent Changes
-
-### 4.0.0 (May 5, 2020)
-
-- Added internationalization support
-  - Added language zh_CN (简体中文)
-  - Language will change automatically according to system locale settings
-- Added support for [Anime4KCPP](https://github.com/TianZerL/Anime4KCPP) in replacement for Anime4K (Java)
-- Driver-specific settings can now be specified in the command line by specifying them after a `--`
-- All driver-specific settings are parsed by the corresponding driver
-- Modularized driver wrappers in Video2X
-- Cleaned up some clutters in the code
-
-### 3.2.0 (April 26, 2020)
-
-- Added support for [SRMD-NCNN-Vulkan](https://github.com/nihui/srmd-ncnn-vulkan)
-
-### 3.1.0 (February 26, 2020)
-
-- Removed the redundant layer of multi-threading since multi-process has to be implemented for launching Windows PE files in sub-processes
-- Added support for graceful exit upon `KeyboardInterrupt` or termination signals
-- Other minor improvements such as replacing `' '.join(execute)` with `shlex.join(execute)`
-
-### Setup Script 1.8.0 (May 5, 2020)
-
-- Added support for Anime4KCPP
-
----
-
-# Full Usage
-
-## General Options:
-
-### -h, --help
-    show this help message and exit
-
-### -i INPUT, --input INPUT
-    source video file/directory
-
-### -o OUTPUT, --output OUTPUT
-    output video file/directory
-
-### -c CONFIG, --config CONFIG
-    video2x config file path
-
-### -d DRIVER, --driver DRIVER
-    upscaling driver (default: waifu2x_caffe)
-
-Available options are:
-
-- waifu2x_caffe
-- waifu2x_converter_cpp
-- waifu2x_ncnn_vulkan
-- srmd_ncnn_vulkan
-- anime4kcpp
-
-### -p PROCESSES, --processes PROCESSES
-    number of processes to use for upscaling (default: 1)
-
-### -v, --version
-    display version, lawful information and exit
-
-## Scaling Options
-
-### --width WIDTH
-    output video width
-
-### --height HEIGHT
-    output video height
-
-### -r RATIO, --ratio RATIO
-    scaling ratio
-
----
+The original clip came from the anime "さくら荘のペットな彼女."\
+Copyright of this clip belongs to 株式会社アニプレックス.
 
 ## License
 
-Licensed under the GNU General Public License Version 3 (GNU GPL v3)
-https://www.gnu.org/licenses/gpl-3.0.txt
+This project is licensed under the [GNU Affero General Public License Version 3 (GNU AGPL v3)](https://www.gnu.org/licenses/agpl-3.0.txt)\
+Copyright (C) 2018-2024 K4YT3X and contributors.
 
-![GPLv3 Icon](https://www.gnu.org/graphics/gplv3-127x51.png)
+![AGPLv3](https://www.gnu.org/graphics/agplv3-155x51.png)
 
-(C) 2018-2020 K4YT3X
+This project includes or depends on these following projects:
 
-## Credits
+| Project                                                                       | License         |
+| ----------------------------------------------------------------------------- | --------------- |
+| [Anime4K](https://github.com/bloc97/Anime4K)                                  | MIT License     |
+| [FFmpeg](https://www.ffmpeg.org/)                                             | LGPLv2.1, GPLv2 |
+| [Real-ESRGAN ncnn Vulkan](https://github.com/xinntao/Real-ESRGAN-ncnn-vulkan) | MIT License     |
+| [ncnn](https://github.com/Tencent/ncnn)                                       | BSD 3-Clause    |
 
-This project relies on the following software and projects.
-
-- [FFmpeg](https://www.ffmpeg.org/)
-- [waifu2x-caffe](https://github.com/lltcggie/waifu2x-caffe)
-- [waifu2x-converter-cpp](https://github.com/DeadSix27/waifu2x-converter-cpp)
-- [waifu2x-ncnn-vulkan](https://github.com/nihui/waifu2x-ncnn-vulkan)
-- [srmd-ncnn-vulkan](https://github.com/nihui/srmd-ncnn-vulkan)
-- [Anime4KCPP](https://github.com/TianZerL/Anime4KCPP)
+More licensing information can be found in the [NOTICE](NOTICE) file.
 
 ## Special Thanks
 
-Appreciations given to the following code contributors:
+Appreciations are given to the following personnel who have contributed significantly to the project.
 
-- @BrianPetkovsek
-- @SAT3LL
+- [@ArchieMeng](https://github.com/archiemeng)
+- [@BrianPetkovsek](https://github.com/BrianPetkovsek)
+- [@ddouglas87](https://github.com/ddouglas87)
+- [@lhanjian](https://github.com/lhanjian)
+- [@nihui](https://github.com/nihui)
+- [@sat3ll](https://github.com/sat3ll)
 
-## Related Projects
+## Similar Projects
 
 - [Dandere2x](https://github.com/CardinalPanda/dandere2x): A lossy video upscaler also built around `waifu2x`, but with video compression techniques to shorten the time needed to process a video.
 - [Waifu2x-Extension-GUI](https://github.com/AaronFeng753/Waifu2x-Extension-GUI): A similar project that focuses more and only on building a better graphical user interface. It is built using C++ and Qt5, and currently only supports the Windows platform.

_config.yml

@@ -1 +0,0 @@
-theme: jekyll-theme-slate

include/conversions.h

@@ -0,0 +1,16 @@
+#ifndef CONVERSIONS_H
+#define CONVERSIONS_H
+
+#include <libavutil/frame.h>
+#include <mat.h>
+
+// Convert AVFrame to another pixel format
+AVFrame *convert_avframe_pix_fmt(AVFrame *src_frame, AVPixelFormat pix_fmt);
+
+// Convert AVFrame to ncnn::Mat
+ncnn::Mat avframe_to_ncnn_mat(AVFrame *frame);
+
+// Convert ncnn::Mat to AVFrame
+AVFrame *ncnn_mat_to_avframe(const ncnn::Mat &mat, AVPixelFormat pix_fmt);
+
+#endif  // CONVERSIONS_H

include/decoder.h

@@ -0,0 +1,15 @@
+#ifndef DECODER_H
+#define DECODER_H
+
+#include <libavcodec/avcodec.h>
+#include <libavfilter/avfilter.h>
+#include <libavformat/avformat.h>
+
+int init_decoder(
+    const char *input_filename,
+    AVFormatContext **fmt_ctx,
+    AVCodecContext **dec_ctx,
+    int *video_stream_index
+);
+
+#endif  // DECODER_H

include/encoder.h

@@ -0,0 +1,21 @@
+#ifndef ENCODER_H
+#define ENCODER_H
+
+#include <libavcodec/avcodec.h>
+#include <libavformat/avformat.h>
+
+#include "libvideo2x.h"
+
+int init_encoder(
+    const char *output_filename,
+    AVFormatContext **ofmt_ctx,
+    AVCodecContext **enc_ctx,
+    AVCodecContext *dec_ctx,
+    EncoderConfig *encoder_config
+);
+
+int encode_and_write_frame(AVFrame *frame, AVCodecContext *enc_ctx, AVFormatContext *ofmt_ctx);
+
+int flush_encoder(AVCodecContext *enc_ctx, AVFormatContext *ofmt_ctx);
+
+#endif  // ENCODER_H

include/filter.h

@@ -0,0 +1,20 @@
+#ifndef FILTER_H
+#define FILTER_H
+
+#include <vector>
+
+extern "C" {
+#include <libavcodec/avcodec.h>
+#include <libavfilter/avfilter.h>
+}
+
+// Abstract base class for filters
+class Filter {
+   public:
+    virtual ~Filter() {}
+    virtual int init(AVCodecContext *dec_ctx, AVCodecContext *enc_ctx) = 0;
+    virtual AVFrame *process_frame(AVFrame *input_frame) = 0;
+    virtual int flush(std::vector<AVFrame *> &processed_frames) = 0;
+};
+
+#endif  // FILTER_H

include/fsutils.h

@@ -0,0 +1,12 @@
+#ifndef FSUTILS_H
+#define FSUTILS_H
+
+#include <filesystem>
+
+bool filepath_is_readable(const std::filesystem::path &path);
+
+std::filesystem::path find_resource_file(const std::filesystem::path &path);
+
+std::string path_to_string(const std::filesystem::path& path);
+
+#endif  // FSUTILS_H

include/getopt.h

@@ -0,0 +1,32 @@
+#ifndef __GETOPT_H__
+#define __GETOPT_H__
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+extern int opterr;   /* if error message should be printed */
+extern int optind;   /* index into parent argv vector */
+extern int optopt;   /* character checked for validity */
+extern int optreset; /* reset getopt */
+extern char *optarg; /* argument associated with option */
+
+struct option {
+    const char *name;
+    int has_arg;
+    int *flag;
+    int val;
+};
+
+#define no_argument 0
+#define required_argument 1
+#define optional_argument 2
+
+int getopt(int, char **, const char *);
+int getopt_long(int, char **, const char *, const struct option *, int *);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* __GETOPT_H__ */

include/libplacebo.h

@@ -0,0 +1,21 @@
+#ifndef PLACEBO_H
+#define PLACEBO_H
+
+#include <filesystem>
+
+#include <libavcodec/avcodec.h>
+#include <libavfilter/avfilter.h>
+#include <libavutil/buffer.h>
+
+int init_libplacebo(
+    AVFilterGraph **filter_graph,
+    AVFilterContext **buffersrc_ctx,
+    AVFilterContext **buffersink_ctx,
+    AVBufferRef **device_ctx,
+    AVCodecContext *dec_ctx,
+    int output_width,
+    int output_height,
+    const std::filesystem::path &shader_path
+);
+
+#endif  // PLACEBO_H

include/libplacebo_filter.h

@@ -0,0 +1,39 @@
+#ifndef LIBPLACEBO_FILTER_H
+#define LIBPLACEBO_FILTER_H
+
+#include <filesystem>
+
+#include <libavutil/buffer.h>
+
+#include "filter.h"
+
+// LibplaceboFilter class definition
+class LibplaceboFilter : public Filter {
+   private:
+    AVFilterGraph *filter_graph;
+    AVFilterContext *buffersrc_ctx;
+    AVFilterContext *buffersink_ctx;
+    AVBufferRef *device_ctx;
+    int output_width;
+    int output_height;
+    const std::filesystem::path shader_path;
+    AVRational output_time_base;
+
+   public:
+    // Constructor
+    LibplaceboFilter(int width, int height, const std::filesystem::path &shader_path);
+
+    // Destructor
+    virtual ~LibplaceboFilter();
+
+    // Initializes the filter with decoder and encoder contexts
+    int init(AVCodecContext *dec_ctx, AVCodecContext *enc_ctx) override;
+
+    // Processes an input frame and returns the processed frame
+    AVFrame *process_frame(AVFrame *input_frame) override;
+
+    // Flushes any remaining frames
+    int flush(std::vector<AVFrame *> &processed_frames) override;
+};
+
+#endif  // LIBPLACEBO_FILTER_H

include/libvideo2x.h

@@ -0,0 +1,86 @@
+#ifndef LIBVIDEO2X_H
+#define LIBVIDEO2X_H
+
+#include <libavutil/pixfmt.h>
+#include <stdint.h>
+#include <time.h>
+
+#include <libavcodec/avcodec.h>
+#include <libavcodec/codec_id.h>
+
+#ifdef _WIN32
+#ifdef LIBVIDEO2X_EXPORTS
+#define LIBVIDEO2X_API __declspec(dllexport)
+#else
+#define LIBVIDEO2X_API __declspec(dllimport)
+#endif
+#else
+#define LIBVIDEO2X_API
+#endif
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+// Enum to specify filter type
+enum FilterType {
+    FILTER_LIBPLACEBO,
+    FILTER_REALESRGAN
+};
+
+// Configuration for Libplacebo filter
+struct LibplaceboConfig {
+    int output_width;
+    int output_height;
+    const char *shader_path;
+};
+
+// Configuration for RealESRGAN filter
+struct RealESRGANConfig {
+    int gpuid;
+    int tta_mode;
+    int scaling_factor;
+    const char *model;
+};
+
+// Unified filter configuration
+struct FilterConfig {
+    enum FilterType filter_type;
+    union {
+        struct LibplaceboConfig libplacebo;
+        struct RealESRGANConfig realesrgan;
+    } config;
+};
+
+// Encoder configuration
+struct EncoderConfig {
+    int output_width;
+    int output_height;
+    enum AVCodecID codec;
+    enum AVPixelFormat pix_fmt;
+    const char *preset;
+    int64_t bit_rate;
+    float crf;
+};
+
+// Processing status
+struct ProcessingStatus {
+    int64_t processed_frames;
+    int64_t total_frames;
+    time_t start_time;
+};
+
+// C-compatible process_video function
+LIBVIDEO2X_API int process_video(
+    const char *input_filename,
+    const char *output_filename,
+    const struct FilterConfig *filter_config,
+    struct EncoderConfig *encoder_config,
+    struct ProcessingStatus *status
+);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif  // LIBVIDEO2X_H

include/realesrgan_filter.h

@@ -0,0 +1,47 @@
+#ifndef REALSRGAN_FILTER_H
+#define REALSRGAN_FILTER_H
+
+#include <filesystem>
+
+#include "filter.h"
+#include "realesrgan.h"
+
+// RealesrganFilter class definition
+class RealesrganFilter : public Filter {
+   private:
+    RealESRGAN *realesrgan;
+    int gpuid;
+    bool tta_mode;
+    int scaling_factor;
+    const char *model;
+    const std::filesystem::path custom_model_param_path;
+    const std::filesystem::path custom_model_bin_path;
+    AVRational input_time_base;
+    AVRational output_time_base;
+    AVPixelFormat output_pix_fmt;
+
+   public:
+    // Constructor
+    RealesrganFilter(
+        int gpuid = 0,
+        bool tta_mode = false,
+        int scaling_factor = 4,
+        const char *model = "realesr-animevideov3",
+        const std::filesystem::path custom_model_bin_pathmodel_param_path = std::filesystem::path(),
+        const std::filesystem::path custom_model_bin_pathmodel_bin_path = std::filesystem::path()
+    );
+
+    // Destructor
+    virtual ~RealesrganFilter();
+
+    // Initializes the filter with decoder and encoder contexts
+    int init(AVCodecContext *dec_ctx, AVCodecContext *enc_ctx) override;
+
+    // Processes an input frame and returns the processed frame
+    AVFrame *process_frame(AVFrame *input_frame) override;
+
+    // Flushes any remaining frames (if necessary)
+    int flush(std::vector<AVFrame *> &processed_frames) override;
+};
+
+#endif

models/realesr-animevideov3-x2.param

@@ -0,0 +1,43 @@
+7767517
+41 42
+Input                    input.1                  0 1 data
+Split                    splitncnn_input0         1 2 data input.1_splitncnn_0 input.1_splitncnn_1
+Convolution              Conv_0                   1 1 input.1_splitncnn_1 54 0=64 1=3 4=1 5=1 6=1728
+PReLU                    PRelu_1                  1 1 54 56 0=64
+Convolution              Conv_2                   1 1 56 57 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_3                  1 1 57 59 0=64
+Convolution              Conv_4                   1 1 59 60 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_5                  1 1 60 62 0=64
+Convolution              Conv_6                   1 1 62 63 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_7                  1 1 63 65 0=64
+Convolution              Conv_8                   1 1 65 66 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_9                  1 1 66 68 0=64
+Convolution              Conv_10                  1 1 68 69 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_11                 1 1 69 71 0=64
+Convolution              Conv_12                  1 1 71 72 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_13                 1 1 72 74 0=64
+Convolution              Conv_14                  1 1 74 75 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_15                 1 1 75 77 0=64
+Convolution              Conv_16                  1 1 77 78 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_17                 1 1 78 80 0=64
+Convolution              Conv_18                  1 1 80 81 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_19                 1 1 81 83 0=64
+Convolution              Conv_20                  1 1 83 84 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_21                 1 1 84 86 0=64
+Convolution              Conv_22                  1 1 86 87 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_23                 1 1 87 89 0=64
+Convolution              Conv_24                  1 1 89 90 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_25                 1 1 90 92 0=64
+Convolution              Conv_26                  1 1 92 93 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_27                 1 1 93 95 0=64
+Convolution              Conv_28                  1 1 95 96 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_29                 1 1 96 98 0=64
+Convolution              Conv_30                  1 1 98 99 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_31                 1 1 99 101 0=64
+Convolution              Conv_32                  1 1 101 102 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_33                 1 1 102 104 0=64
+Convolution              Conv_34                  1 1 104 105 0=48 1=3 4=1 5=1 6=27648
+PixelShuffle             DepthToSpace_35          1 1 105 106 0=4
+Interp                   Resize_37                1 1 input.1_splitncnn_0 111 0=1 1=4.000000e+00 2=4.000000e+00
+BinaryOp                 Add_38                   2 1 106 111 112
+Interp                   Resize_40                1 1 112 output 0=3 1=5.000000e-01 2=5.000000e-01

models/realesr-animevideov3-x3.param

@@ -0,0 +1,43 @@
+7767517
+41 42
+Input                    input.1                  0 1 data
+Split                    splitncnn_input0         1 2 data input.1_splitncnn_0 input.1_splitncnn_1
+Convolution              Conv_0                   1 1 input.1_splitncnn_1 54 0=64 1=3 4=1 5=1 6=1728
+PReLU                    PRelu_1                  1 1 54 56 0=64
+Convolution              Conv_2                   1 1 56 57 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_3                  1 1 57 59 0=64
+Convolution              Conv_4                   1 1 59 60 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_5                  1 1 60 62 0=64
+Convolution              Conv_6                   1 1 62 63 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_7                  1 1 63 65 0=64
+Convolution              Conv_8                   1 1 65 66 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_9                  1 1 66 68 0=64
+Convolution              Conv_10                  1 1 68 69 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_11                 1 1 69 71 0=64
+Convolution              Conv_12                  1 1 71 72 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_13                 1 1 72 74 0=64
+Convolution              Conv_14                  1 1 74 75 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_15                 1 1 75 77 0=64
+Convolution              Conv_16                  1 1 77 78 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_17                 1 1 78 80 0=64
+Convolution              Conv_18                  1 1 80 81 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_19                 1 1 81 83 0=64
+Convolution              Conv_20                  1 1 83 84 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_21                 1 1 84 86 0=64
+Convolution              Conv_22                  1 1 86 87 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_23                 1 1 87 89 0=64
+Convolution              Conv_24                  1 1 89 90 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_25                 1 1 90 92 0=64
+Convolution              Conv_26                  1 1 92 93 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_27                 1 1 93 95 0=64
+Convolution              Conv_28                  1 1 95 96 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_29                 1 1 96 98 0=64
+Convolution              Conv_30                  1 1 98 99 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_31                 1 1 99 101 0=64
+Convolution              Conv_32                  1 1 101 102 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_33                 1 1 102 104 0=64
+Convolution              Conv_34                  1 1 104 105 0=48 1=3 4=1 5=1 6=27648
+PixelShuffle             DepthToSpace_35          1 1 105 106 0=4
+Interp                   Resize_37                1 1 input.1_splitncnn_0 111 0=1 1=4.000000e+00 2=4.000000e+00
+BinaryOp                 Add_38                   2 1 106 111 112
+Interp                   Resize_40                1 1 112 output 0=3 1=7.500000e-01 2=7.500000e-01

models/realesr-animevideov3-x4.param

@@ -0,0 +1,42 @@
+7767517
+40 41
+Input                    input.1                  0 1 data
+Split                    splitncnn_input0         1 2 data input.1_splitncnn_0 input.1_splitncnn_1
+Convolution              Conv_0                   1 1 input.1_splitncnn_1 54 0=64 1=3 4=1 5=1 6=1728
+PReLU                    PRelu_1                  1 1 54 56 0=64
+Convolution              Conv_2                   1 1 56 57 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_3                  1 1 57 59 0=64
+Convolution              Conv_4                   1 1 59 60 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_5                  1 1 60 62 0=64
+Convolution              Conv_6                   1 1 62 63 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_7                  1 1 63 65 0=64
+Convolution              Conv_8                   1 1 65 66 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_9                  1 1 66 68 0=64
+Convolution              Conv_10                  1 1 68 69 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_11                 1 1 69 71 0=64
+Convolution              Conv_12                  1 1 71 72 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_13                 1 1 72 74 0=64
+Convolution              Conv_14                  1 1 74 75 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_15                 1 1 75 77 0=64
+Convolution              Conv_16                  1 1 77 78 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_17                 1 1 78 80 0=64
+Convolution              Conv_18                  1 1 80 81 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_19                 1 1 81 83 0=64
+Convolution              Conv_20                  1 1 83 84 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_21                 1 1 84 86 0=64
+Convolution              Conv_22                  1 1 86 87 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_23                 1 1 87 89 0=64
+Convolution              Conv_24                  1 1 89 90 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_25                 1 1 90 92 0=64
+Convolution              Conv_26                  1 1 92 93 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_27                 1 1 93 95 0=64
+Convolution              Conv_28                  1 1 95 96 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_29                 1 1 96 98 0=64
+Convolution              Conv_30                  1 1 98 99 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_31                 1 1 99 101 0=64
+Convolution              Conv_32                  1 1 101 102 0=64 1=3 4=1 5=1 6=36864
+PReLU                    PRelu_33                 1 1 102 104 0=64
+Convolution              Conv_34                  1 1 104 105 0=48 1=3 4=1 5=1 6=27648
+PixelShuffle             DepthToSpace_35          1 1 105 106 0=4
+Interp                   Resize_37                1 1 input.1_splitncnn_0 111 0=1 1=4.000000e+00 2=4.000000e+00
+BinaryOp                 Add_38                   2 1 106 111 output

models/realesrgan-plus-anime-x4.param

@@ -0,0 +1,270 @@
+7767517
+268 473
+Input                    input.1                  0 1 data
+Convolution              Conv_0                   1 1 data 193 0=64 1=3 4=1 5=1 6=1728
+Split                    splitncnn_0              1 8 193 193_splitncnn_0 193_splitncnn_1 193_splitncnn_2 193_splitncnn_3 193_splitncnn_4 193_splitncnn_5 193_splitncnn_6 193_splitncnn_7
+Convolution              Conv_1                   1 1 193_splitncnn_7 195 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_1              1 4 195 195_splitncnn_0 195_splitncnn_1 195_splitncnn_2 195_splitncnn_3
+Concat                   Concat_3                 2 1 193_splitncnn_6 195_splitncnn_3 196
+Convolution              Conv_4                   1 1 196 198 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_2              1 3 198 198_splitncnn_0 198_splitncnn_1 198_splitncnn_2
+Concat                   Concat_6                 3 1 193_splitncnn_5 195_splitncnn_2 198_splitncnn_2 199
+Convolution              Conv_7                   1 1 199 201 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_3              1 2 201 201_splitncnn_0 201_splitncnn_1
+Concat                   Concat_9                 4 1 193_splitncnn_4 195_splitncnn_1 198_splitncnn_1 201_splitncnn_1 202
+Convolution              Conv_10                  1 1 202 204 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_12                5 1 193_splitncnn_3 195_splitncnn_0 198_splitncnn_0 201_splitncnn_0 204 205
+Convolution              Conv_13                  1 1 205 206 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_16                   2 1 206 193_splitncnn_2 209 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_4              1 6 209 209_splitncnn_0 209_splitncnn_1 209_splitncnn_2 209_splitncnn_3 209_splitncnn_4 209_splitncnn_5
+Convolution              Conv_17                  1 1 209_splitncnn_5 211 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_5              1 4 211 211_splitncnn_0 211_splitncnn_1 211_splitncnn_2 211_splitncnn_3
+Concat                   Concat_19                2 1 209_splitncnn_4 211_splitncnn_3 212
+Convolution              Conv_20                  1 1 212 214 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_6              1 3 214 214_splitncnn_0 214_splitncnn_1 214_splitncnn_2
+Concat                   Concat_22                3 1 209_splitncnn_3 211_splitncnn_2 214_splitncnn_2 215
+Convolution              Conv_23                  1 1 215 217 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_7              1 2 217 217_splitncnn_0 217_splitncnn_1
+Concat                   Concat_25                4 1 209_splitncnn_2 211_splitncnn_1 214_splitncnn_1 217_splitncnn_1 218
+Convolution              Conv_26                  1 1 218 220 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_28                5 1 209_splitncnn_1 211_splitncnn_0 214_splitncnn_0 217_splitncnn_0 220 221
+Convolution              Conv_29                  1 1 221 222 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_32                   2 1 222 209_splitncnn_0 225 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_8              1 6 225 225_splitncnn_0 225_splitncnn_1 225_splitncnn_2 225_splitncnn_3 225_splitncnn_4 225_splitncnn_5
+Convolution              Conv_33                  1 1 225_splitncnn_5 227 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_9              1 4 227 227_splitncnn_0 227_splitncnn_1 227_splitncnn_2 227_splitncnn_3
+Concat                   Concat_35                2 1 225_splitncnn_4 227_splitncnn_3 228
+Convolution              Conv_36                  1 1 228 230 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_10             1 3 230 230_splitncnn_0 230_splitncnn_1 230_splitncnn_2
+Concat                   Concat_38                3 1 225_splitncnn_3 227_splitncnn_2 230_splitncnn_2 231
+Convolution              Conv_39                  1 1 231 233 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_11             1 2 233 233_splitncnn_0 233_splitncnn_1
+Concat                   Concat_41                4 1 225_splitncnn_2 227_splitncnn_1 230_splitncnn_1 233_splitncnn_1 234
+Convolution              Conv_42                  1 1 234 236 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_44                5 1 225_splitncnn_1 227_splitncnn_0 230_splitncnn_0 233_splitncnn_0 236 237
+Convolution              Conv_45                  1 1 237 238 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_48                   2 1 238 225_splitncnn_0 241 0=1 -23301=2,2.000000e-01,1.000000e+00
+Eltwise                  Add_51                   2 1 241 193_splitncnn_1 244 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_12             1 7 244 244_splitncnn_0 244_splitncnn_1 244_splitncnn_2 244_splitncnn_3 244_splitncnn_4 244_splitncnn_5 244_splitncnn_6
+Convolution              Conv_52                  1 1 244_splitncnn_6 246 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_13             1 4 246 246_splitncnn_0 246_splitncnn_1 246_splitncnn_2 246_splitncnn_3
+Concat                   Concat_54                2 1 244_splitncnn_5 246_splitncnn_3 247
+Convolution              Conv_55                  1 1 247 249 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_14             1 3 249 249_splitncnn_0 249_splitncnn_1 249_splitncnn_2
+Concat                   Concat_57                3 1 244_splitncnn_4 246_splitncnn_2 249_splitncnn_2 250
+Convolution              Conv_58                  1 1 250 252 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_15             1 2 252 252_splitncnn_0 252_splitncnn_1
+Concat                   Concat_60                4 1 244_splitncnn_3 246_splitncnn_1 249_splitncnn_1 252_splitncnn_1 253
+Convolution              Conv_61                  1 1 253 255 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_63                5 1 244_splitncnn_2 246_splitncnn_0 249_splitncnn_0 252_splitncnn_0 255 256
+Convolution              Conv_64                  1 1 256 257 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_67                   2 1 257 244_splitncnn_1 260 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_16             1 6 260 260_splitncnn_0 260_splitncnn_1 260_splitncnn_2 260_splitncnn_3 260_splitncnn_4 260_splitncnn_5
+Convolution              Conv_68                  1 1 260_splitncnn_5 262 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_17             1 4 262 262_splitncnn_0 262_splitncnn_1 262_splitncnn_2 262_splitncnn_3
+Concat                   Concat_70                2 1 260_splitncnn_4 262_splitncnn_3 263
+Convolution              Conv_71                  1 1 263 265 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_18             1 3 265 265_splitncnn_0 265_splitncnn_1 265_splitncnn_2
+Concat                   Concat_73                3 1 260_splitncnn_3 262_splitncnn_2 265_splitncnn_2 266
+Convolution              Conv_74                  1 1 266 268 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_19             1 2 268 268_splitncnn_0 268_splitncnn_1
+Concat                   Concat_76                4 1 260_splitncnn_2 262_splitncnn_1 265_splitncnn_1 268_splitncnn_1 269
+Convolution              Conv_77                  1 1 269 271 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_79                5 1 260_splitncnn_1 262_splitncnn_0 265_splitncnn_0 268_splitncnn_0 271 272
+Convolution              Conv_80                  1 1 272 273 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_83                   2 1 273 260_splitncnn_0 276 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_20             1 6 276 276_splitncnn_0 276_splitncnn_1 276_splitncnn_2 276_splitncnn_3 276_splitncnn_4 276_splitncnn_5
+Convolution              Conv_84                  1 1 276_splitncnn_5 278 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_21             1 4 278 278_splitncnn_0 278_splitncnn_1 278_splitncnn_2 278_splitncnn_3
+Concat                   Concat_86                2 1 276_splitncnn_4 278_splitncnn_3 279
+Convolution              Conv_87                  1 1 279 281 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_22             1 3 281 281_splitncnn_0 281_splitncnn_1 281_splitncnn_2
+Concat                   Concat_89                3 1 276_splitncnn_3 278_splitncnn_2 281_splitncnn_2 282
+Convolution              Conv_90                  1 1 282 284 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_23             1 2 284 284_splitncnn_0 284_splitncnn_1
+Concat                   Concat_92                4 1 276_splitncnn_2 278_splitncnn_1 281_splitncnn_1 284_splitncnn_1 285
+Convolution              Conv_93                  1 1 285 287 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_95                5 1 276_splitncnn_1 278_splitncnn_0 281_splitncnn_0 284_splitncnn_0 287 288
+Convolution              Conv_96                  1 1 288 289 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_99                   2 1 289 276_splitncnn_0 292 0=1 -23301=2,2.000000e-01,1.000000e+00
+Eltwise                  Add_102                  2 1 292 244_splitncnn_0 295 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_24             1 7 295 295_splitncnn_0 295_splitncnn_1 295_splitncnn_2 295_splitncnn_3 295_splitncnn_4 295_splitncnn_5 295_splitncnn_6
+Convolution              Conv_103                 1 1 295_splitncnn_6 297 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_25             1 4 297 297_splitncnn_0 297_splitncnn_1 297_splitncnn_2 297_splitncnn_3
+Concat                   Concat_105               2 1 295_splitncnn_5 297_splitncnn_3 298
+Convolution              Conv_106                 1 1 298 300 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_26             1 3 300 300_splitncnn_0 300_splitncnn_1 300_splitncnn_2
+Concat                   Concat_108               3 1 295_splitncnn_4 297_splitncnn_2 300_splitncnn_2 301
+Convolution              Conv_109                 1 1 301 303 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_27             1 2 303 303_splitncnn_0 303_splitncnn_1
+Concat                   Concat_111               4 1 295_splitncnn_3 297_splitncnn_1 300_splitncnn_1 303_splitncnn_1 304
+Convolution              Conv_112                 1 1 304 306 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_114               5 1 295_splitncnn_2 297_splitncnn_0 300_splitncnn_0 303_splitncnn_0 306 307
+Convolution              Conv_115                 1 1 307 308 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_118                  2 1 308 295_splitncnn_1 311 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_28             1 6 311 311_splitncnn_0 311_splitncnn_1 311_splitncnn_2 311_splitncnn_3 311_splitncnn_4 311_splitncnn_5
+Convolution              Conv_119                 1 1 311_splitncnn_5 313 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_29             1 4 313 313_splitncnn_0 313_splitncnn_1 313_splitncnn_2 313_splitncnn_3
+Concat                   Concat_121               2 1 311_splitncnn_4 313_splitncnn_3 314
+Convolution              Conv_122                 1 1 314 316 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_30             1 3 316 316_splitncnn_0 316_splitncnn_1 316_splitncnn_2
+Concat                   Concat_124               3 1 311_splitncnn_3 313_splitncnn_2 316_splitncnn_2 317
+Convolution              Conv_125                 1 1 317 319 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_31             1 2 319 319_splitncnn_0 319_splitncnn_1
+Concat                   Concat_127               4 1 311_splitncnn_2 313_splitncnn_1 316_splitncnn_1 319_splitncnn_1 320
+Convolution              Conv_128                 1 1 320 322 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_130               5 1 311_splitncnn_1 313_splitncnn_0 316_splitncnn_0 319_splitncnn_0 322 323
+Convolution              Conv_131                 1 1 323 324 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_134                  2 1 324 311_splitncnn_0 327 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_32             1 6 327 327_splitncnn_0 327_splitncnn_1 327_splitncnn_2 327_splitncnn_3 327_splitncnn_4 327_splitncnn_5
+Convolution              Conv_135                 1 1 327_splitncnn_5 329 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_33             1 4 329 329_splitncnn_0 329_splitncnn_1 329_splitncnn_2 329_splitncnn_3
+Concat                   Concat_137               2 1 327_splitncnn_4 329_splitncnn_3 330
+Convolution              Conv_138                 1 1 330 332 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_34             1 3 332 332_splitncnn_0 332_splitncnn_1 332_splitncnn_2
+Concat                   Concat_140               3 1 327_splitncnn_3 329_splitncnn_2 332_splitncnn_2 333
+Convolution              Conv_141                 1 1 333 335 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_35             1 2 335 335_splitncnn_0 335_splitncnn_1
+Concat                   Concat_143               4 1 327_splitncnn_2 329_splitncnn_1 332_splitncnn_1 335_splitncnn_1 336
+Convolution              Conv_144                 1 1 336 338 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_146               5 1 327_splitncnn_1 329_splitncnn_0 332_splitncnn_0 335_splitncnn_0 338 339
+Convolution              Conv_147                 1 1 339 340 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_150                  2 1 340 327_splitncnn_0 343 0=1 -23301=2,2.000000e-01,1.000000e+00
+Eltwise                  Add_153                  2 1 343 295_splitncnn_0 346 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_36             1 7 346 346_splitncnn_0 346_splitncnn_1 346_splitncnn_2 346_splitncnn_3 346_splitncnn_4 346_splitncnn_5 346_splitncnn_6
+Convolution              Conv_154                 1 1 346_splitncnn_6 348 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_37             1 4 348 348_splitncnn_0 348_splitncnn_1 348_splitncnn_2 348_splitncnn_3
+Concat                   Concat_156               2 1 346_splitncnn_5 348_splitncnn_3 349
+Convolution              Conv_157                 1 1 349 351 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_38             1 3 351 351_splitncnn_0 351_splitncnn_1 351_splitncnn_2
+Concat                   Concat_159               3 1 346_splitncnn_4 348_splitncnn_2 351_splitncnn_2 352
+Convolution              Conv_160                 1 1 352 354 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_39             1 2 354 354_splitncnn_0 354_splitncnn_1
+Concat                   Concat_162               4 1 346_splitncnn_3 348_splitncnn_1 351_splitncnn_1 354_splitncnn_1 355
+Convolution              Conv_163                 1 1 355 357 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_165               5 1 346_splitncnn_2 348_splitncnn_0 351_splitncnn_0 354_splitncnn_0 357 358
+Convolution              Conv_166                 1 1 358 359 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_169                  2 1 359 346_splitncnn_1 362 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_40             1 6 362 362_splitncnn_0 362_splitncnn_1 362_splitncnn_2 362_splitncnn_3 362_splitncnn_4 362_splitncnn_5
+Convolution              Conv_170                 1 1 362_splitncnn_5 364 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_41             1 4 364 364_splitncnn_0 364_splitncnn_1 364_splitncnn_2 364_splitncnn_3
+Concat                   Concat_172               2 1 362_splitncnn_4 364_splitncnn_3 365
+Convolution              Conv_173                 1 1 365 367 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_42             1 3 367 367_splitncnn_0 367_splitncnn_1 367_splitncnn_2
+Concat                   Concat_175               3 1 362_splitncnn_3 364_splitncnn_2 367_splitncnn_2 368
+Convolution              Conv_176                 1 1 368 370 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_43             1 2 370 370_splitncnn_0 370_splitncnn_1
+Concat                   Concat_178               4 1 362_splitncnn_2 364_splitncnn_1 367_splitncnn_1 370_splitncnn_1 371
+Convolution              Conv_179                 1 1 371 373 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_181               5 1 362_splitncnn_1 364_splitncnn_0 367_splitncnn_0 370_splitncnn_0 373 374
+Convolution              Conv_182                 1 1 374 375 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_185                  2 1 375 362_splitncnn_0 378 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_44             1 6 378 378_splitncnn_0 378_splitncnn_1 378_splitncnn_2 378_splitncnn_3 378_splitncnn_4 378_splitncnn_5
+Convolution              Conv_186                 1 1 378_splitncnn_5 380 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_45             1 4 380 380_splitncnn_0 380_splitncnn_1 380_splitncnn_2 380_splitncnn_3
+Concat                   Concat_188               2 1 378_splitncnn_4 380_splitncnn_3 381
+Convolution              Conv_189                 1 1 381 383 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_46             1 3 383 383_splitncnn_0 383_splitncnn_1 383_splitncnn_2
+Concat                   Concat_191               3 1 378_splitncnn_3 380_splitncnn_2 383_splitncnn_2 384
+Convolution              Conv_192                 1 1 384 386 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_47             1 2 386 386_splitncnn_0 386_splitncnn_1
+Concat                   Concat_194               4 1 378_splitncnn_2 380_splitncnn_1 383_splitncnn_1 386_splitncnn_1 387
+Convolution              Conv_195                 1 1 387 389 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_197               5 1 378_splitncnn_1 380_splitncnn_0 383_splitncnn_0 386_splitncnn_0 389 390
+Convolution              Conv_198                 1 1 390 391 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_201                  2 1 391 378_splitncnn_0 394 0=1 -23301=2,2.000000e-01,1.000000e+00
+Eltwise                  Add_204                  2 1 394 346_splitncnn_0 397 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_48             1 7 397 397_splitncnn_0 397_splitncnn_1 397_splitncnn_2 397_splitncnn_3 397_splitncnn_4 397_splitncnn_5 397_splitncnn_6
+Convolution              Conv_205                 1 1 397_splitncnn_6 399 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_49             1 4 399 399_splitncnn_0 399_splitncnn_1 399_splitncnn_2 399_splitncnn_3
+Concat                   Concat_207               2 1 397_splitncnn_5 399_splitncnn_3 400
+Convolution              Conv_208                 1 1 400 402 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_50             1 3 402 402_splitncnn_0 402_splitncnn_1 402_splitncnn_2
+Concat                   Concat_210               3 1 397_splitncnn_4 399_splitncnn_2 402_splitncnn_2 403
+Convolution              Conv_211                 1 1 403 405 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_51             1 2 405 405_splitncnn_0 405_splitncnn_1
+Concat                   Concat_213               4 1 397_splitncnn_3 399_splitncnn_1 402_splitncnn_1 405_splitncnn_1 406
+Convolution              Conv_214                 1 1 406 408 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_216               5 1 397_splitncnn_2 399_splitncnn_0 402_splitncnn_0 405_splitncnn_0 408 409
+Convolution              Conv_217                 1 1 409 410 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_220                  2 1 410 397_splitncnn_1 413 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_52             1 6 413 413_splitncnn_0 413_splitncnn_1 413_splitncnn_2 413_splitncnn_3 413_splitncnn_4 413_splitncnn_5
+Convolution              Conv_221                 1 1 413_splitncnn_5 415 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_53             1 4 415 415_splitncnn_0 415_splitncnn_1 415_splitncnn_2 415_splitncnn_3
+Concat                   Concat_223               2 1 413_splitncnn_4 415_splitncnn_3 416
+Convolution              Conv_224                 1 1 416 418 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_54             1 3 418 418_splitncnn_0 418_splitncnn_1 418_splitncnn_2
+Concat                   Concat_226               3 1 413_splitncnn_3 415_splitncnn_2 418_splitncnn_2 419
+Convolution              Conv_227                 1 1 419 421 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_55             1 2 421 421_splitncnn_0 421_splitncnn_1
+Concat                   Concat_229               4 1 413_splitncnn_2 415_splitncnn_1 418_splitncnn_1 421_splitncnn_1 422
+Convolution              Conv_230                 1 1 422 424 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_232               5 1 413_splitncnn_1 415_splitncnn_0 418_splitncnn_0 421_splitncnn_0 424 425
+Convolution              Conv_233                 1 1 425 426 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_236                  2 1 426 413_splitncnn_0 429 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_56             1 6 429 429_splitncnn_0 429_splitncnn_1 429_splitncnn_2 429_splitncnn_3 429_splitncnn_4 429_splitncnn_5
+Convolution              Conv_237                 1 1 429_splitncnn_5 431 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_57             1 4 431 431_splitncnn_0 431_splitncnn_1 431_splitncnn_2 431_splitncnn_3
+Concat                   Concat_239               2 1 429_splitncnn_4 431_splitncnn_3 432
+Convolution              Conv_240                 1 1 432 434 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_58             1 3 434 434_splitncnn_0 434_splitncnn_1 434_splitncnn_2
+Concat                   Concat_242               3 1 429_splitncnn_3 431_splitncnn_2 434_splitncnn_2 435
+Convolution              Conv_243                 1 1 435 437 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_59             1 2 437 437_splitncnn_0 437_splitncnn_1
+Concat                   Concat_245               4 1 429_splitncnn_2 431_splitncnn_1 434_splitncnn_1 437_splitncnn_1 438
+Convolution              Conv_246                 1 1 438 440 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_248               5 1 429_splitncnn_1 431_splitncnn_0 434_splitncnn_0 437_splitncnn_0 440 441
+Convolution              Conv_249                 1 1 441 442 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_252                  2 1 442 429_splitncnn_0 445 0=1 -23301=2,2.000000e-01,1.000000e+00
+Eltwise                  Add_255                  2 1 445 397_splitncnn_0 448 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_60             1 7 448 448_splitncnn_0 448_splitncnn_1 448_splitncnn_2 448_splitncnn_3 448_splitncnn_4 448_splitncnn_5 448_splitncnn_6
+Convolution              Conv_256                 1 1 448_splitncnn_6 450 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_61             1 4 450 450_splitncnn_0 450_splitncnn_1 450_splitncnn_2 450_splitncnn_3
+Concat                   Concat_258               2 1 448_splitncnn_5 450_splitncnn_3 451
+Convolution              Conv_259                 1 1 451 453 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_62             1 3 453 453_splitncnn_0 453_splitncnn_1 453_splitncnn_2
+Concat                   Concat_261               3 1 448_splitncnn_4 450_splitncnn_2 453_splitncnn_2 454
+Convolution              Conv_262                 1 1 454 456 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_63             1 2 456 456_splitncnn_0 456_splitncnn_1
+Concat                   Concat_264               4 1 448_splitncnn_3 450_splitncnn_1 453_splitncnn_1 456_splitncnn_1 457
+Convolution              Conv_265                 1 1 457 459 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_267               5 1 448_splitncnn_2 450_splitncnn_0 453_splitncnn_0 456_splitncnn_0 459 460
+Convolution              Conv_268                 1 1 460 461 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_271                  2 1 461 448_splitncnn_1 464 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_64             1 6 464 464_splitncnn_0 464_splitncnn_1 464_splitncnn_2 464_splitncnn_3 464_splitncnn_4 464_splitncnn_5
+Convolution              Conv_272                 1 1 464_splitncnn_5 466 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_65             1 4 466 466_splitncnn_0 466_splitncnn_1 466_splitncnn_2 466_splitncnn_3
+Concat                   Concat_274               2 1 464_splitncnn_4 466_splitncnn_3 467
+Convolution              Conv_275                 1 1 467 469 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_66             1 3 469 469_splitncnn_0 469_splitncnn_1 469_splitncnn_2
+Concat                   Concat_277               3 1 464_splitncnn_3 466_splitncnn_2 469_splitncnn_2 470
+Convolution              Conv_278                 1 1 470 472 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_67             1 2 472 472_splitncnn_0 472_splitncnn_1
+Concat                   Concat_280               4 1 464_splitncnn_2 466_splitncnn_1 469_splitncnn_1 472_splitncnn_1 473
+Convolution              Conv_281                 1 1 473 475 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_283               5 1 464_splitncnn_1 466_splitncnn_0 469_splitncnn_0 472_splitncnn_0 475 476
+Convolution              Conv_284                 1 1 476 477 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_287                  2 1 477 464_splitncnn_0 480 0=1 -23301=2,2.000000e-01,1.000000e+00
+Split                    splitncnn_68             1 6 480 480_splitncnn_0 480_splitncnn_1 480_splitncnn_2 480_splitncnn_3 480_splitncnn_4 480_splitncnn_5
+Convolution              Conv_288                 1 1 480_splitncnn_5 482 0=32 1=3 4=1 5=1 6=18432 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_69             1 4 482 482_splitncnn_0 482_splitncnn_1 482_splitncnn_2 482_splitncnn_3
+Concat                   Concat_290               2 1 480_splitncnn_4 482_splitncnn_3 483
+Convolution              Conv_291                 1 1 483 485 0=32 1=3 4=1 5=1 6=27648 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_70             1 3 485 485_splitncnn_0 485_splitncnn_1 485_splitncnn_2
+Concat                   Concat_293               3 1 480_splitncnn_3 482_splitncnn_2 485_splitncnn_2 486
+Convolution              Conv_294                 1 1 486 488 0=32 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Split                    splitncnn_71             1 2 488 488_splitncnn_0 488_splitncnn_1
+Concat                   Concat_296               4 1 480_splitncnn_2 482_splitncnn_1 485_splitncnn_1 488_splitncnn_1 489
+Convolution              Conv_297                 1 1 489 491 0=32 1=3 4=1 5=1 6=46080 9=2 -23310=1,2.000000e-01
+Concat                   Concat_299               5 1 480_splitncnn_1 482_splitncnn_0 485_splitncnn_0 488_splitncnn_0 491 492
+Convolution              Conv_300                 1 1 492 493 0=64 1=3 4=1 5=1 6=110592
+Eltwise                  Add_303                  2 1 493 480_splitncnn_0 496 0=1 -23301=2,2.000000e-01,1.000000e+00
+Eltwise                  Add_306                  2 1 496 448_splitncnn_0 499 0=1 -23301=2,2.000000e-01,1.000000e+00
+Convolution              Conv_307                 1 1 499 500 0=64 1=3 4=1 5=1 6=36864
+BinaryOp                 Add_308                  2 1 193_splitncnn_0 500 501
+Interp                   Resize_310               1 1 501 506 0=1 1=2.000000e+00 2=2.000000e+00
+Convolution              Conv_311                 1 1 506 508 0=64 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Interp                   Resize_314               1 1 508 513 0=1 1=2.000000e+00 2=2.000000e+00
+Convolution              Conv_315                 1 1 513 515 0=64 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Convolution              Conv_317                 1 1 515 517 0=64 1=3 4=1 5=1 6=36864 9=2 -23310=1,2.000000e-01
+Convolution              Conv_319                 1 1 517 output 0=3 1=3 4=1 5=1 6=1728

scripts/download_anime4k_glsl.py

@@ -0,0 +1,46 @@
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+import shutil
+from pathlib import Path
+
+import requests
+
+ANIME4K_COMMIT = "master"
+GITHUB_GLSL_ROOT = (
+    f"https://raw.githubusercontent.com/bloc97/Anime4K/{ANIME4K_COMMIT}/glsl"
+)
+SHADERS_DIR = Path(__file__).parent.parent / "data"
+
+
+def download_and_combine_files():
+
+    modes = {
+        "ModeA": [
+            f"{GITHUB_GLSL_ROOT}/Restore/Anime4K_Clamp_Highlights.glsl",
+            f"{GITHUB_GLSL_ROOT}/Restore/Anime4K_Restore_CNN_VL.glsl",
+            f"{GITHUB_GLSL_ROOT}/Upscale/Anime4K_Upscale_CNN_x2_VL.glsl",
+            f"{GITHUB_GLSL_ROOT}/Upscale/Anime4K_AutoDownscalePre_x2.glsl",
+            f"{GITHUB_GLSL_ROOT}/Upscale/Anime4K_AutoDownscalePre_x4.glsl",
+            f"{GITHUB_GLSL_ROOT}/Upscale/Anime4K_Upscale_CNN_x2_M.glsl",
+        ]
+    }
+
+    for mode in modes:
+        file_contents = ""
+        for file in modes[mode]:
+            response = requests.get(file, timeout=5)
+            response.raise_for_status()
+            file_contents += response.text + "\n"
+
+        with (SHADERS_DIR / Path(f"Anime4K_{mode}.glsl")).open("w") as output_file:
+            output_file.write(file_contents)
+
+
+if __name__ == "__main__":
+    # clear shaders directory
+    if SHADERS_DIR.exists():
+        shutil.rmtree(SHADERS_DIR)
+    SHADERS_DIR.mkdir(exist_ok=True)
+
+    # download and combine shaders
+    download_and_combine_files()

src/build.ps1

@@ -1,92 +0,0 @@
-<#
-Name: Video2X Build Script
-Creator: K4YT3X
-Date Created: May 6, 2020
-Last Modified: May 7, 2020
-
-Description: A PowerShell script that will build Video2X
-executable (PE) releases automatically using PyInstaller.
-This script is currently only tuned for K4YT3X's environment.
-
-To start a PowerShell session with execution policy bypass
-powershell –ExecutionPolicy Bypass
-#>
-
-if ($args.count -ne 1){
-    Write-Host -ForegroundColor White "Usage:`n .\build.ps1 VIDEO2X_VERSION"
-    Exit
-}
-
-# version number
-$SCRIPT_VERSION = "1.0.1"
-$VIDEO2X_VERSION = $args[0]
-
-Write-Host -ForegroundColor White "Video2X Building Script Version $($SCRIPT_VERSION)
-Starting to build Video2X release packages
-Building Video2X release $($VIDEO2X_VERSION)"
-
-# build Video2X CLI
-Write-Host -ForegroundColor White "`nBuilding Video2X CLI"
-pyinstaller --noconfirm --log-level=WARN `
-    --onefile `
-    --add-data="wrappers;wrappers" `
-    --icon="images\video2x.ico" `
-    video2x.py
-
-# build Video2X GUI
-Write-Host -ForegroundColor White "`nBuilding Video2X GUI"
-pyinstaller --noconfirm --log-level=WARN `
-    --onefile `
-    --add-data="images;images" `
-    --add-data="locale;locale" `
-    --add-data="video2x_gui.ui;." `
-    --add-data="wrappers;wrappers" `
-    --icon="images\video2x.ico" `
-    video2x_gui.py
-
-# build setup script
-Write-Host -ForegroundColor White "`nBuilding Video2X setup script"
-pyinstaller --noconfirm --log-level=WARN `
-    --onefile `
-    --icon="images\video2x.ico" `
-    video2x_setup.py
-
-# remove old builds if found
-if (Test-Path "$($VIDEO2X_VERSION)" -PathType any) {
-    Remove-Item -path "$($VIDEO2X_VERSION)" -recurse
-}
-
-# create build directory
-New-Item "$($VIDEO2X_VERSION)" -ItemType Directory
-
-# copy files into corresponding builds
-# full edition
-Write-Host -ForegroundColor White "`nCreating full package"
-New-Item "$($VIDEO2X_VERSION)\video2x-$($VIDEO2X_VERSION)-win32-full" -ItemType Directory
-Copy-Item "dist\video2x.exe" -Destination "$($VIDEO2X_VERSION)\video2x-$($VIDEO2X_VERSION)-win32-full\"
-Copy-Item "dist\video2x_gui.exe" -Destination "$($VIDEO2X_VERSION)\video2x-$($VIDEO2X_VERSION)-win32-full\"
-Copy-Item -Path "$env:LOCALAPPDATA\video2x" -Destination "$($VIDEO2X_VERSION)\video2x-$($VIDEO2X_VERSION)-win32-full\dependencies" -Recurse
-
-# overwrite paths to relative paths
-(Get-Content "video2x.yaml").replace("%LOCALAPPDATA%\video2x", "dependencies") | Set-Content "video2x.yaml.relative"
-Move-Item "video2x.yaml.relative" -Destination "$($VIDEO2X_VERSION)\video2x-$($VIDEO2X_VERSION)-win32-full\video2x.yaml"
-
-# light edition
-Write-Host -ForegroundColor White "`nCreating light package"
-New-Item "$($VIDEO2X_VERSION)\video2x-$($VIDEO2X_VERSION)-win32-light" -ItemType Directory
-Copy-Item "dist\video2x.exe" -Destination "$($VIDEO2X_VERSION)\video2x-$($VIDEO2X_VERSION)-win32-light\"
-Copy-Item "dist\video2x_gui.exe" -Destination "$($VIDEO2X_VERSION)\video2x-$($VIDEO2X_VERSION)-win32-light\"
-Copy-Item "dist\video2x_setup.exe" -Destination "$($VIDEO2X_VERSION)\video2x-$($VIDEO2X_VERSION)-win32-light\"
-Copy-Item "video2x.yaml" -Destination "$($VIDEO2X_VERSION)\video2x-$($VIDEO2X_VERSION)-win32-light\"
-Copy-Item "requirements.txt" -Destination "$($VIDEO2X_VERSION)\video2x-$($VIDEO2X_VERSION)-win32-light\"
-
-# clean up temporary files
-Write-Host -ForegroundColor White "`nDeleting temporary files"
-$pathsToRemove = "__pycache__", "build", "dist", "*.spec"
-
-foreach ($path in $pathsToRemove){
-    Write-Host "Removing path: $($path)"
-    Remove-Item -path $path -recurse
-}
-
-Write-Host -ForegroundColor White "`nBuild script finished"

src/conversions.cpp

@@ -0,0 +1,202 @@
+#include <cstdio>
+
+// FFmpeg includes
+extern "C" {
+#include <libavutil/frame.h>
+#include <libavutil/imgutils.h>
+#include <libswscale/swscale.h>
+}
+
+// ncnn includes
+#include <mat.h>
+
+#include "conversions.h"
+
+// Convert AVFrame format
+AVFrame *convert_avframe_pix_fmt(AVFrame *src_frame, AVPixelFormat pix_fmt) {
+    AVFrame *dst_frame = av_frame_alloc();
+    if (dst_frame == nullptr) {
+        fprintf(stderr, "Failed to allocate destination AVFrame.\n");
+        return nullptr;
+    }
+
+    dst_frame->format = pix_fmt;
+    dst_frame->width = src_frame->width;
+    dst_frame->height = src_frame->height;
+
+    // Allocate memory for the converted frame
+    if (av_frame_get_buffer(dst_frame, 32) < 0) {
+        fprintf(stderr, "Failed to allocate memory for AVFrame.\n");
+        av_frame_free(&dst_frame);
+        return nullptr;
+    }
+
+    // Create a SwsContext for pixel format conversion
+    SwsContext *sws_ctx = sws_getContext(
+        src_frame->width,
+        src_frame->height,
+        static_cast<AVPixelFormat>(src_frame->format),
+        dst_frame->width,
+        dst_frame->height,
+        pix_fmt,
+        SWS_BILINEAR,
+        nullptr,
+        nullptr,
+        nullptr
+    );
+
+    if (sws_ctx == nullptr) {
+        fprintf(stderr, "Failed to initialize swscale context.\n");
+        av_frame_free(&dst_frame);
+        return nullptr;
+    }
+
+    // Perform the conversion
+    sws_scale(
+        sws_ctx,
+        src_frame->data,
+        src_frame->linesize,
+        0,
+        src_frame->height,
+        dst_frame->data,
+        dst_frame->linesize
+    );
+
+    // Clean up
+    sws_freeContext(sws_ctx);
+
+    return dst_frame;
+}
+
+// Convert AVFrame to ncnn::Mat by copying the data
+ncnn::Mat avframe_to_ncnn_mat(AVFrame *frame) {
+    AVFrame *converted_frame = nullptr;
+
+    // Convert to BGR24 format if necessary
+    if (frame->format != AV_PIX_FMT_BGR24) {
+        converted_frame = convert_avframe_pix_fmt(frame, AV_PIX_FMT_BGR24);
+        if (!converted_frame) {
+            fprintf(stderr, "Failed to convert AVFrame to BGR24.\n");
+            return ncnn::Mat();  // Return an empty ncnn::Mat on failure
+        }
+    } else {
+        converted_frame = frame;  // If the frame is already in BGR24, use it directly
+    }
+
+    // Allocate a new ncnn::Mat and copy the data
+    int width = converted_frame->width;
+    int height = converted_frame->height;
+    ncnn::Mat ncnn_image = ncnn::Mat(width, height, (size_t)3, 3);  // BGR has 3 channels
+
+    // Manually copy the pixel data from AVFrame to the new ncnn::Mat
+    const uint8_t *src_data = converted_frame->data[0];
+    for (int y = 0; y < height; y++) {
+        uint8_t *dst_row = ncnn_image.row<uint8_t>(y);
+        const uint8_t *src_row = src_data + y * converted_frame->linesize[0];
+        memcpy(dst_row, src_row, width * 3);  // Copy 3 channels (BGR) per pixel
+    }
+
+    // If we allocated a converted frame, free it
+    if (converted_frame != frame) {
+        av_frame_free(&converted_frame);
+    }
+
+    return ncnn_image;
+}
+
+// Convert ncnn::Mat to AVFrame with a specified pixel format (this part is unchanged)
+AVFrame *ncnn_mat_to_avframe(const ncnn::Mat &mat, AVPixelFormat pix_fmt) {
+    int ret;
+
+    // Step 1: Allocate a destination AVFrame for the specified pixel format
+    AVFrame *dst_frame = av_frame_alloc();
+    if (!dst_frame) {
+        fprintf(stderr, "Failed to allocate destination AVFrame.\n");
+        return nullptr;
+    }
+
+    dst_frame->format = pix_fmt;
+    dst_frame->width = mat.w;
+    dst_frame->height = mat.h;
+
+    // Allocate memory for the frame buffer
+    if (av_frame_get_buffer(dst_frame, 32) < 0) {
+        fprintf(stderr, "Failed to allocate memory for destination AVFrame.\n");
+        av_frame_free(&dst_frame);
+        return nullptr;
+    }
+
+    // Step 2: Convert ncnn::Mat to BGR AVFrame
+    AVFrame *bgr_frame = av_frame_alloc();
+    if (!bgr_frame) {
+        fprintf(stderr, "Failed to allocate intermediate BGR AVFrame.\n");
+        av_frame_free(&dst_frame);
+        return nullptr;
+    }
+
+    bgr_frame->format = AV_PIX_FMT_BGR24;
+    bgr_frame->width = mat.w;
+    bgr_frame->height = mat.h;
+
+    // Allocate memory for the intermediate BGR frame
+    if (av_frame_get_buffer(bgr_frame, 32) < 0) {
+        fprintf(stderr, "Failed to allocate memory for BGR AVFrame.\n");
+        av_frame_free(&dst_frame);
+        av_frame_free(&bgr_frame);
+        return nullptr;
+    }
+
+    // Copy data from ncnn::Mat to the BGR AVFrame
+    // mat.to_pixels(bgr_frame->data[0], ncnn::Mat::PIXEL_BGR);
+
+    // Manually copy the pixel data from ncnn::Mat to the BGR AVFrame
+    for (int y = 0; y < mat.h; y++) {
+        uint8_t *dst_row = bgr_frame->data[0] + y * bgr_frame->linesize[0];
+        const uint8_t *src_row = mat.row<const uint8_t>(y);
+        memcpy(dst_row, src_row, mat.w * 3);  // Copy 3 channels (BGR) per pixel
+    }
+
+    // Step 3: Convert the BGR frame to the desired pixel format
+    SwsContext *sws_ctx = sws_getContext(
+        bgr_frame->width,
+        bgr_frame->height,
+        AV_PIX_FMT_BGR24,
+        dst_frame->width,
+        dst_frame->height,
+        pix_fmt,
+        SWS_BILINEAR,
+        nullptr,
+        nullptr,
+        nullptr
+    );
+
+    if (sws_ctx == nullptr) {
+        fprintf(stderr, "Failed to initialize swscale context.\n");
+        av_frame_free(&bgr_frame);
+        av_frame_free(&dst_frame);
+        return nullptr;
+    }
+
+    // Perform the conversion
+    ret = sws_scale(
+        sws_ctx,
+        bgr_frame->data,
+        bgr_frame->linesize,
+        0,
+        bgr_frame->height,
+        dst_frame->data,
+        dst_frame->linesize
+    );
+
+    // Clean up
+    sws_freeContext(sws_ctx);
+    av_frame_free(&bgr_frame);
+
+    if (ret != dst_frame->height) {
+        fprintf(stderr, "Failed to convert BGR AVFrame to destination pixel format.\n");
+        av_frame_free(&dst_frame);
+        return nullptr;
+    }
+
+    return dst_frame;
+}

src/decoder.cpp

@@ -0,0 +1,79 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+extern "C" {
+#include <libavcodec/avcodec.h>
+#include <libavfilter/avfilter.h>
+#include <libavfilter/buffersink.h>
+#include <libavfilter/buffersrc.h>
+#include <libavformat/avformat.h>
+#include <libavutil/opt.h>
+#include <libavutil/pixdesc.h>
+#include <libavutil/rational.h>
+}
+
+int init_decoder(
+    const char *input_filename,
+    AVFormatContext **fmt_ctx,
+    AVCodecContext **dec_ctx,
+    int *video_stream_index
+) {
+    AVFormatContext *ifmt_ctx = NULL;
+    AVCodecContext *codec_ctx = NULL;
+    int ret;
+
+    if ((ret = avformat_open_input(&ifmt_ctx, input_filename, NULL, NULL)) < 0) {
+        fprintf(stderr, "Could not open input file '%s'\n", input_filename);
+        return ret;
+    }
+
+    if ((ret = avformat_find_stream_info(ifmt_ctx, NULL)) < 0) {
+        fprintf(stderr, "Failed to retrieve input stream information\n");
+        return ret;
+    }
+
+    // Find the first video stream
+    ret = av_find_best_stream(ifmt_ctx, AVMEDIA_TYPE_VIDEO, -1, -1, NULL, 0);
+    if (ret < 0) {
+        fprintf(stderr, "Could not find video stream in the input, aborting\n");
+        return ret;
+    }
+
+    int stream_index = ret;
+    AVStream *video_stream = ifmt_ctx->streams[stream_index];
+
+    // Set up the decoder
+    const AVCodec *dec = avcodec_find_decoder(video_stream->codecpar->codec_id);
+    if (!dec) {
+        fprintf(stderr, "Failed to find decoder for stream #%u\n", stream_index);
+        return AVERROR_DECODER_NOT_FOUND;
+    }
+
+    codec_ctx = avcodec_alloc_context3(dec);
+    if (!codec_ctx) {
+        fprintf(stderr, "Failed to allocate the decoder context\n");
+        return AVERROR(ENOMEM);
+    }
+
+    if ((ret = avcodec_parameters_to_context(codec_ctx, video_stream->codecpar)) < 0) {
+        fprintf(stderr, "Failed to copy decoder parameters to input decoder context\n");
+        return ret;
+    }
+
+    // Set decoder time base and frame rate
+    codec_ctx->time_base = video_stream->time_base;
+    codec_ctx->pkt_timebase = video_stream->time_base;
+    codec_ctx->framerate = av_guess_frame_rate(ifmt_ctx, video_stream, NULL);
+
+    if ((ret = avcodec_open2(codec_ctx, dec, NULL)) < 0) {
+        fprintf(stderr, "Failed to open decoder for stream #%u\n", stream_index);
+        return ret;
+    }
+
+    *fmt_ctx = ifmt_ctx;
+    *dec_ctx = codec_ctx;
+    *video_stream_index = stream_index;
+
+    return 0;
+}

src/encoder.cpp

@@ -0,0 +1,206 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+extern "C" {
+#include <libavcodec/avcodec.h>
+#include <libavcodec/codec.h>
+#include <libavcodec/codec_id.h>
+#include <libavfilter/avfilter.h>
+#include <libavfilter/buffersink.h>
+#include <libavfilter/buffersrc.h>
+#include <libavformat/avformat.h>
+#include <libavutil/opt.h>
+#include <libavutil/pixdesc.h>
+#include <libavutil/rational.h>
+}
+
+#include "conversions.h"
+#include "libvideo2x.h"
+
+int init_encoder(
+    const char *output_filename,
+    AVFormatContext **ofmt_ctx,
+    AVCodecContext **enc_ctx,
+    AVCodecContext *dec_ctx,
+    EncoderConfig *encoder_config
+) {
+    AVFormatContext *fmt_ctx = NULL;
+    AVCodecContext *codec_ctx = NULL;
+    int ret;
+
+    avformat_alloc_output_context2(&fmt_ctx, NULL, NULL, output_filename);
+    if (!fmt_ctx) {
+        fprintf(stderr, "Could not create output context\n");
+        return AVERROR_UNKNOWN;
+    }
+
+    // Create a new video stream
+    const AVCodec *enc = avcodec_find_encoder(encoder_config->codec);
+    if (!enc) {
+        fprintf(stderr, "Necessary encoder not found\n");
+        return AVERROR_ENCODER_NOT_FOUND;
+    }
+
+    AVStream *out_stream = avformat_new_stream(fmt_ctx, NULL);
+    if (!out_stream) {
+        fprintf(stderr, "Failed allocating output stream\n");
+        return AVERROR_UNKNOWN;
+    }
+
+    codec_ctx = avcodec_alloc_context3(enc);
+    if (!codec_ctx) {
+        fprintf(stderr, "Failed to allocate the encoder context\n");
+        return AVERROR(ENOMEM);
+    }
+
+    // Set encoding parameters
+    codec_ctx->height = encoder_config->output_height;
+    codec_ctx->width = encoder_config->output_width;
+    codec_ctx->sample_aspect_ratio = dec_ctx->sample_aspect_ratio;
+    codec_ctx->pix_fmt = encoder_config->pix_fmt;
+    codec_ctx->time_base = av_inv_q(dec_ctx->framerate);
+
+    if (codec_ctx->time_base.num == 0 || codec_ctx->time_base.den == 0) {
+        codec_ctx->time_base = av_inv_q(av_guess_frame_rate(fmt_ctx, out_stream, NULL));
+    }
+
+    // Set the bit rate and other encoder parameters if needed
+    codec_ctx->bit_rate = encoder_config->bit_rate;
+    codec_ctx->gop_size = 60;     // Keyframe interval
+    codec_ctx->max_b_frames = 3;  // B-frames
+    codec_ctx->keyint_min = 60;   // Maximum GOP size
+
+    char crf_str[16];
+    snprintf(crf_str, sizeof(crf_str), "%.f", encoder_config->crf);
+    if (encoder_config->codec == AV_CODEC_ID_H264 || encoder_config->codec == AV_CODEC_ID_HEVC) {
+        av_opt_set(codec_ctx->priv_data, "crf", crf_str, 0);
+        av_opt_set(codec_ctx->priv_data, "preset", encoder_config->preset, 0);
+    }
+
+    if (fmt_ctx->oformat->flags & AVFMT_GLOBALHEADER) {
+        codec_ctx->flags |= AV_CODEC_FLAG_GLOBAL_HEADER;
+    }
+
+    if ((ret = avcodec_open2(codec_ctx, enc, NULL)) < 0) {
+        fprintf(stderr, "Cannot open video encoder\n");
+        return ret;
+    }
+
+    ret = avcodec_parameters_from_context(out_stream->codecpar, codec_ctx);
+    if (ret < 0) {
+        fprintf(stderr, "Failed to copy encoder parameters to output stream\n");
+        return ret;
+    }
+
+    out_stream->time_base = codec_ctx->time_base;
+
+    // Open the output file
+    if (!(fmt_ctx->oformat->flags & AVFMT_NOFILE)) {
+        ret = avio_open(&fmt_ctx->pb, output_filename, AVIO_FLAG_WRITE);
+        if (ret < 0) {
+            fprintf(stderr, "Could not open output file '%s'\n", output_filename);
+            return ret;
+        }
+    }
+
+    *ofmt_ctx = fmt_ctx;
+    *enc_ctx = codec_ctx;
+
+    return 0;
+}
+
+int encode_and_write_frame(AVFrame *frame, AVCodecContext *enc_ctx, AVFormatContext *ofmt_ctx) {
+    int ret;
+
+    // Convert the frame to the encoder's pixel format if needed
+    if (frame->format != enc_ctx->pix_fmt) {
+        AVFrame *converted_frame = convert_avframe_pix_fmt(frame, enc_ctx->pix_fmt);
+        if (!converted_frame) {
+            fprintf(stderr, "Error converting frame to encoder's pixel format\n");
+            return AVERROR_EXTERNAL;
+        }
+
+        converted_frame->pts = frame->pts;
+        frame = converted_frame;
+    }
+
+    AVPacket *enc_pkt = av_packet_alloc();
+    if (!enc_pkt) {
+        fprintf(stderr, "Could not allocate AVPacket\n");
+        return AVERROR(ENOMEM);
+    }
+
+    ret = avcodec_send_frame(enc_ctx, frame);
+    if (ret < 0) {
+        fprintf(stderr, "Error sending frame to encoder\n");
+        av_packet_free(&enc_pkt);
+        return ret;
+    }
+
+    while (ret >= 0) {
+        ret = avcodec_receive_packet(enc_ctx, enc_pkt);
+        if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {
+            av_packet_unref(enc_pkt);
+            break;
+        } else if (ret < 0) {
+            fprintf(stderr, "Error during encoding\n");
+            av_packet_free(&enc_pkt);
+            return ret;
+        }
+
+        // Rescale packet timestamps
+        av_packet_rescale_ts(enc_pkt, enc_ctx->time_base, ofmt_ctx->streams[0]->time_base);
+        enc_pkt->stream_index = ofmt_ctx->streams[0]->index;
+
+        // Write the packet
+        ret = av_interleaved_write_frame(ofmt_ctx, enc_pkt);
+        av_packet_unref(enc_pkt);
+        if (ret < 0) {
+            fprintf(stderr, "Error muxing packet\n");
+            av_packet_free(&enc_pkt);
+            return ret;
+        }
+    }
+
+    av_packet_free(&enc_pkt);
+    return 0;
+}
+
+int flush_encoder(AVCodecContext *enc_ctx, AVFormatContext *ofmt_ctx) {
+    int ret;
+    AVPacket *enc_pkt = av_packet_alloc();
+    if (!enc_pkt) {
+        fprintf(stderr, "Could not allocate AVPacket\n");
+        return AVERROR(ENOMEM);
+    }
+
+    ret = avcodec_send_frame(enc_ctx, NULL);
+    while (ret >= 0) {
+        ret = avcodec_receive_packet(enc_ctx, enc_pkt);
+        if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {
+            av_packet_unref(enc_pkt);
+            break;
+        } else if (ret < 0) {
+            fprintf(stderr, "Error during encoding\n");
+            av_packet_free(&enc_pkt);
+            return ret;
+        }
+
+        // Rescale packet timestamps
+        av_packet_rescale_ts(enc_pkt, enc_ctx->time_base, ofmt_ctx->streams[0]->time_base);
+        enc_pkt->stream_index = ofmt_ctx->streams[0]->index;
+
+        // Write the packet
+        ret = av_interleaved_write_frame(ofmt_ctx, enc_pkt);
+        av_packet_unref(enc_pkt);
+        if (ret < 0) {
+            fprintf(stderr, "Error muxing packet\n");
+            av_packet_free(&enc_pkt);
+            return ret;
+        }
+    }
+
+    av_packet_free(&enc_pkt);
+    return 0;
+}

src/exceptions.py

@@ -1,28 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Name: Video2X Exceptions
-Dev: K4YT3X
-Date Created: December 13, 2018
-Last Modified: July 27, 2019
-"""
-
-
-class ArgumentError(Exception):
-    def __init__(self, message):
-        super().__init__(message)
-
-
-class StreamNotFoundError(Exception):
-    def __init__(self, message):
-        super().__init__(message)
-
-
-class UnrecognizedDriverError(Exception):
-    def __init__(self, message):
-        super().__init__(message)
-
-
-class UnsupportedPixelError(Exception):
-    def __init__(self, message):
-        super().__init__(message)

src/fsutils.cpp

@@ -0,0 +1,94 @@
+#include <filesystem>
+
+#if _WIN32
+#include <windows.h>
+#include <cwchar>
+#else
+#include <unistd.h>
+#include <cstring>
+#endif
+
+#include "fsutils.h"
+
+#if _WIN32
+std::filesystem::path get_executable_directory() {
+    std::vector<wchar_t> filepath(MAX_PATH);
+
+    // Get the executable path, expanding the buffer if necessary
+    DWORD size = GetModuleFileNameW(NULL, filepath.data(), static_cast<DWORD>(filepath.size()));
+    if (size == 0) {
+        fprintf(stderr, "Error getting executable path: %lu\n", GetLastError());
+        return std::filesystem::path();
+    }
+
+    // Resize the buffer if necessary
+    while (size >= filepath.size()) {
+        filepath.resize(filepath.size() * 2);
+        size = GetModuleFileNameW(NULL, filepath.data(), static_cast<DWORD>(filepath.size()));
+        if (size == 0) {
+            fprintf(stderr, "Error getting executable path: %lu\n", GetLastError());
+            return std::filesystem::path();
+        }
+    }
+
+    // Create a std::filesystem::path from the filepath and return its parent path
+    std::filesystem::path execpath(filepath.data());
+    return execpath.parent_path();
+}
+#else   // _WIN32
+std::filesystem::path get_executable_directory() {
+    std::error_code ec;
+    std::filesystem::path filepath = std::filesystem::read_symlink("/proc/self/exe", ec);
+
+    if (ec) {
+        fprintf(stderr, "Error reading /proc/self/exe: %s\n", ec.message().c_str());
+        return std::filesystem::path();
+    }
+
+    return filepath.parent_path();
+}
+#endif  // _WIN32
+
+bool filepath_is_readable(const std::filesystem::path &path) {
+#if _WIN32
+    FILE *fp = _wfopen(path.c_str(), L"rb");
+#else   // _WIN32
+    FILE *fp = fopen(path.c_str(), "rb");
+#endif  // _WIN32
+    if (!fp) {
+        return false;
+    }
+
+    fclose(fp);
+    return true;
+}
+
+std::filesystem::path find_resource_file(const std::filesystem::path &path) {
+    if (filepath_is_readable(path)) {
+        return path;
+    }
+
+    if (filepath_is_readable(std::filesystem::path("/usr/share/video2x/") / path)) {
+        return std::filesystem::path("/usr/share/video2x/") / path;
+    }
+
+    return get_executable_directory() / path;
+}
+
+std::string path_to_string(const std::filesystem::path &path) {
+#if _WIN32
+    std::wstring wide_path = path.wstring();
+    int buffer_size =
+        WideCharToMultiByte(CP_UTF8, 0, wide_path.c_str(), -1, nullptr, 0, nullptr, nullptr);
+    if (buffer_size == 0) {
+        return std::string();
+    }
+    std::vector<char> buffer(buffer_size);
+    WideCharToMultiByte(
+        CP_UTF8, 0, wide_path.c_str(), -1, buffer.data(), buffer_size, nullptr, nullptr
+    );
+    return std::string(buffer.data());
+#else
+    return path.string();
+#endif
+}

src/getopt.c

@@ -0,0 +1,249 @@
+/*
+ * Copyright (c) 1987, 1993, 1994, 1996
+ *	The Regents of the University of California.  All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the names of the copyright holders nor the names of its
+ *    contributors may be used to endorse or promote products derived from
+ *    this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS ``AS
+ * IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
+ * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+ * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+#include <assert.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "getopt.h"
+
+/*
+extern int opterr;
+extern int optind;
+extern int optopt;
+extern int optreset;
+extern char *optarg;
+*/
+
+int opterr = 1;      /* if error message should be printed */
+int optind = 1;      /* index into parent argv vector */
+int optopt = 0;      /* character checked for validity */
+int optreset = 0;    /* reset getopt */
+char *optarg = NULL; /* argument associated with option */
+
+#ifndef __P
+#define __P(x) x
+#endif
+#define _DIAGASSERT(x) assert(x)
+
+static char *__progname __P((char *));
+int getopt_internal __P((int, char *const *, const char *));
+
+static char *__progname(char *nargv0) {
+    char *tmp;
+
+    _DIAGASSERT(nargv0 != NULL);
+
+    tmp = strrchr(nargv0, '/');
+    if (tmp) {
+        tmp++;
+    } else {
+        tmp = nargv0;
+    }
+    return (tmp);
+}
+
+#define BADCH (int)'?'
+#define BADARG (int)':'
+#define EMSG ""
+
+/*
+ * getopt --
+ *	Parse argc/argv argument vector.
+ */
+int getopt_internal(int nargc, char *const *nargv, const char *ostr) {
+    static char *place = EMSG; /* option letter processing */
+    char *oli;                 /* option letter list index */
+
+    _DIAGASSERT(nargv != NULL);
+    _DIAGASSERT(ostr != NULL);
+
+    if (optreset || !*place) { /* update scanning pointer */
+        optreset = 0;
+        if (optind >= nargc || *(place = nargv[optind]) != '-') {
+            place = EMSG;
+            return (-1);
+        }
+        if (place[1] && *++place == '-') { /* found "--" */
+            /* ++optind; */
+            place = EMSG;
+            return (-2);
+        }
+    } /* option letter okay? */
+    if ((optopt = (int)*place++) == (int)':' || !(oli = strchr(ostr, optopt))) {
+        /*
+         * if the user didn't specify '-' as an option,
+         * assume it means -1.
+         */
+        if (optopt == (int)'-') {
+            return (-1);
+        }
+        if (!*place) {
+            ++optind;
+        }
+        if (opterr && *ostr != ':') {
+            (void)fprintf(stderr, "%s: illegal option -- %c\n", __progname(nargv[0]), optopt);
+        }
+        return (BADCH);
+    }
+    if (*++oli != ':') { /* don't need argument */
+        optarg = NULL;
+        if (!*place) {
+            ++optind;
+        }
+    } else {          /* need an argument */
+        if (*place) { /* no white space */
+            optarg = place;
+        } else if (nargc <= ++optind) { /* no arg */
+            place = EMSG;
+            if ((opterr) && (*ostr != ':')) {
+                (void)fprintf(
+                    stderr, "%s: option requires an argument -- %c\n", __progname(nargv[0]), optopt
+                );
+            }
+            return (BADARG);
+        } else { /* white space */
+            optarg = nargv[optind];
+        }
+        place = EMSG;
+        ++optind;
+    }
+    return (optopt); /* dump back option letter */
+}
+
+#if 0
+/*
+ * getopt --
+ *	Parse argc/argv argument vector.
+ */
+int
+getopt2(nargc, nargv, ostr)
+	int nargc;
+	char * const *nargv;
+	const char *ostr;
+{
+	int retval;
+
+	if ((retval = getopt_internal(nargc, nargv, ostr)) == -2) {
+		retval = -1;
+		++optind; 
+	}
+	return(retval);
+}
+#endif
+
+/*
+ * getopt_long --
+ *	Parse argc/argv argument vector.
+ */
+int getopt_long(
+    int nargc,
+    char **nargv,
+    const char *options,
+    const struct option *long_options,
+    int *index
+) {
+    int retval;
+
+    _DIAGASSERT(nargv != NULL);
+    _DIAGASSERT(options != NULL);
+    _DIAGASSERT(long_options != NULL);
+    /* index may be NULL */
+
+    if ((retval = getopt_internal(nargc, nargv, options)) == -2) {
+        char *current_argv = nargv[optind++] + 2, *has_equal;
+        int i, match = -1;
+        size_t current_argv_len;
+
+        if (*current_argv == '\0') {
+            return (-1);
+        }
+        if ((has_equal = strchr(current_argv, '=')) != NULL) {
+            current_argv_len = has_equal - current_argv;
+            has_equal++;
+        } else {
+            current_argv_len = strlen(current_argv);
+        }
+
+        for (i = 0; long_options[i].name; i++) {
+            if (strncmp(current_argv, long_options[i].name, current_argv_len)) {
+                continue;
+            }
+
+            if (strlen(long_options[i].name) == current_argv_len) {
+                match = i;
+                break;
+            }
+            if (match == -1) {
+                match = i;
+            }
+        }
+        if (match != -1) {
+            if (long_options[match].has_arg == required_argument ||
+                long_options[match].has_arg == optional_argument) {
+                if (has_equal) {
+                    optarg = has_equal;
+                } else {
+                    optarg = nargv[optind++];
+                }
+            }
+            if ((long_options[match].has_arg == required_argument) && (optarg == NULL)) {
+                /*
+                 * Missing argument, leading :
+                 * indicates no error should be generated
+                 */
+                if ((opterr) && (*options != ':')) {
+                    (void)fprintf(
+                        stderr,
+                        "%s: option requires an argument -- %s\n",
+                        __progname(nargv[0]),
+                        current_argv
+                    );
+                }
+                return (BADARG);
+            }
+        } else { /* No matching argument */
+            if ((opterr) && (*options != ':')) {
+                (void
+                )fprintf(stderr, "%s: illegal option -- %s\n", __progname(nargv[0]), current_argv);
+            }
+            return (BADCH);
+        }
+        if (long_options[match].flag) {
+            *long_options[match].flag = long_options[match].val;
+            retval = 0;
+        } else {
+            retval = long_options[match].val;
+        }
+        if (index) {
+            *index = match;
+        }
+    }
+    return (retval);
+}

src/image_cleaner.py

@@ -1,79 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Name: Video2X Image Cleaner
-Author: BrianPetkovsek
-Date Created: March 24, 2019
-Last Modified: July 27, 2019
-
-Editor: K4YT3X
-Last Modified: March 23, 2020
-
-Editor: 28598519a
-Last Modified: March 23, 2020
-
-Description: This class is to remove the extracted frames
-that have already been upscaled.
-"""
-
-# built-in imports
-import threading
-import time
-
-
-class ImageCleaner(threading.Thread):
-    """ Video2X Image Cleaner
-
-    This class creates an object that keeps track of extracted
-    frames that has already been upscaled and are not needed
-    anymore. It then deletes them to save disk space.
-
-    Extends:
-        threading.Thread
-    """
-
-    def __init__(self, input_directory, output_directory, threads):
-        threading.Thread.__init__(self)
-        self.input_directory = input_directory
-        self.output_directory = output_directory
-        self.threads = threads
-        self.running = False
-
-    def run(self):
-        """ Run image cleaner
-        """
-        self.running = True
-
-        while self.running:
-            self.remove_upscaled_frames()
-            time.sleep(1)
-
-    def stop(self):
-        """ Stop the image cleaner
-        """
-        self.running = False
-        self.join()
-
-    def remove_upscaled_frames(self):
-        """ remove frames that have already been upscaled
-
-        This method compares the files in the extracted frames
-        directory with the upscaled frames directory, and removes
-        the frames that has already been upscaled.
-        """
-
-        # list all images in the extracted frames
-        output_frames = [f.name for f in self.output_directory.iterdir() if f.is_file()]
-
-        # compare and remove frames downscaled images that finished being upscaled
-        # within each thread's  extracted frames directory
-        for thread_id in range(self.threads):
-            dir_path = self.input_directory / str(thread_id)
-
-            # for each file within all the directories
-            for file in dir_path.iterdir():
-                # if file also exists in the output directory, then the file
-                # has already been processed, thus not needed anymore
-                if file.is_file() and file.name in output_frames:
-                    file.unlink()
-                    output_frames.remove(file.name)

src/libplacebo.cpp

@@ -0,0 +1,161 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <filesystem>
+
+extern "C" {
+#include <libavcodec/avcodec.h>
+#include <libavfilter/avfilter.h>
+#include <libavfilter/buffersink.h>
+#include <libavfilter/buffersrc.h>
+#include <libavformat/avformat.h>
+#include <libavutil/buffer.h>
+#include <libavutil/hwcontext.h>
+#include <libavutil/opt.h>
+#include <libavutil/pixdesc.h>
+#include <libavutil/rational.h>
+#include <libswscale/swscale.h>
+}
+
+#include "fsutils.h"
+
+int init_libplacebo(
+    AVFilterGraph **filter_graph,
+    AVFilterContext **buffersrc_ctx,
+    AVFilterContext **buffersink_ctx,
+    AVBufferRef **device_ctx,
+    AVCodecContext *dec_ctx,
+    int output_width,
+    int output_height,
+    const std::filesystem::path &shader_path
+) {
+    char args[512];
+    int ret;
+
+    // Initialize the Vulkan hardware device
+    AVBufferRef *hw_device_ctx = av_hwdevice_ctx_alloc(AV_HWDEVICE_TYPE_VULKAN);
+    ret = av_hwdevice_ctx_create(&hw_device_ctx, AV_HWDEVICE_TYPE_VULKAN, NULL, NULL, 0);
+    if (ret < 0) {
+        fprintf(stderr, "Unable to initialize Vulkan device\n");
+        return ret;
+    }
+
+    AVFilterGraph *graph = avfilter_graph_alloc();
+    if (!graph) {
+        fprintf(stderr, "Unable to create filter graph.\n");
+        return AVERROR(ENOMEM);
+    }
+
+    // Create buffer source
+    const AVFilter *buffersrc = avfilter_get_by_name("buffer");
+    snprintf(
+        args,
+        sizeof(args),
+        "video_size=%dx%d:pix_fmt=%d:time_base=%d/%d:frame_rate=%d/%d:"
+        "pixel_aspect=%d/%d:colorspace=%d",
+        dec_ctx->width,
+        dec_ctx->height,
+        dec_ctx->pix_fmt,
+        dec_ctx->time_base.num,
+        dec_ctx->time_base.den,
+        dec_ctx->framerate.num,
+        dec_ctx->framerate.den,
+        dec_ctx->sample_aspect_ratio.num,
+        dec_ctx->sample_aspect_ratio.den,
+        dec_ctx->colorspace
+    );
+
+    ret = avfilter_graph_create_filter(buffersrc_ctx, buffersrc, "in", args, NULL, graph);
+    if (ret < 0) {
+        fprintf(stderr, "Cannot create buffer source\n");
+        av_buffer_unref(&hw_device_ctx);
+        avfilter_graph_free(&graph);
+        return ret;
+    }
+
+    AVFilterContext *last_filter = *buffersrc_ctx;
+
+    // Create the libplacebo filter
+    const AVFilter *libplacebo_filter = avfilter_get_by_name("libplacebo");
+    if (!libplacebo_filter) {
+        fprintf(stderr, "Filter 'libplacebo' not found\n");
+        av_buffer_unref(&hw_device_ctx);
+        avfilter_graph_free(&graph);
+        return AVERROR_FILTER_NOT_FOUND;
+    }
+
+    // Convert the shader path to a string since filter args is const char *
+    std::string shader_path_string = path_to_string(shader_path);
+
+#ifdef _WIN32
+    // libplacebo does not recognize the Windows '\\' path separator
+    std::replace(shader_path_string.begin(), shader_path_string.end(), '\\', '/');
+#endif
+
+    // Prepare the filter arguments
+    char filter_args[512];
+    snprintf(
+        filter_args,
+        sizeof(filter_args),
+        "w=%d:h=%d:upscaler=ewa_lanczos:custom_shader_path=%s",
+        output_width,
+        output_height,
+        shader_path_string.c_str()
+    );
+
+    AVFilterContext *libplacebo_ctx;
+    ret = avfilter_graph_create_filter(
+        &libplacebo_ctx, libplacebo_filter, "libplacebo", filter_args, NULL, graph
+    );
+    if (ret < 0) {
+        fprintf(stderr, "Cannot create libplacebo filter\n");
+        av_buffer_unref(&hw_device_ctx);
+        avfilter_graph_free(&graph);
+        return ret;
+    }
+
+    // Set the hardware device context to Vulkan
+    libplacebo_ctx->hw_device_ctx = av_buffer_ref(hw_device_ctx);
+
+    // Link buffersrc to libplacebo
+    ret = avfilter_link(last_filter, 0, libplacebo_ctx, 0);
+    if (ret < 0) {
+        fprintf(stderr, "Error connecting buffersrc to libplacebo filter\n");
+        av_buffer_unref(&hw_device_ctx);
+        avfilter_graph_free(&graph);
+        return ret;
+    }
+
+    last_filter = libplacebo_ctx;
+
+    // Create buffer sink
+    const AVFilter *buffersink = avfilter_get_by_name("buffersink");
+    ret = avfilter_graph_create_filter(buffersink_ctx, buffersink, "out", NULL, NULL, graph);
+    if (ret < 0) {
+        fprintf(stderr, "Cannot create buffer sink\n");
+        av_buffer_unref(&hw_device_ctx);
+        avfilter_graph_free(&graph);
+        return ret;
+    }
+
+    // Link libplacebo to buffersink
+    ret = avfilter_link(last_filter, 0, *buffersink_ctx, 0);
+    if (ret < 0) {
+        fprintf(stderr, "Error connecting libplacebo filter to buffersink\n");
+        av_buffer_unref(&hw_device_ctx);
+        avfilter_graph_free(&graph);
+        return ret;
+    }
+
+    // Configure the filter graph
+    ret = avfilter_graph_config(graph, NULL);
+    if (ret < 0) {
+        fprintf(stderr, "Error configuring the filter graph\n");
+        av_buffer_unref(&hw_device_ctx);
+        avfilter_graph_free(&graph);
+        return ret;
+    }
+
+    *filter_graph = graph;
+    *device_ctx = hw_device_ctx;
+    return 0;
+}

src/libplacebo_filter.cpp

@@ -0,0 +1,139 @@
+#include <cstdio>
+
+extern "C" {
+#include <libavcodec/avcodec.h>
+#include <libavfilter/buffersink.h>
+#include <libavfilter/buffersrc.h>
+#include <libavutil/buffer.h>
+}
+
+#include "fsutils.h"
+#include "libplacebo.h"
+#include "libplacebo_filter.h"
+
+LibplaceboFilter::LibplaceboFilter(int width, int height, const std::filesystem::path &shader_path)
+    : filter_graph(nullptr),
+      buffersrc_ctx(nullptr),
+      buffersink_ctx(nullptr),
+      device_ctx(nullptr),
+      output_width(width),
+      output_height(height),
+      shader_path(std::move(shader_path)) {}
+
+LibplaceboFilter::~LibplaceboFilter() {
+    if (buffersrc_ctx) {
+        avfilter_free(buffersrc_ctx);
+        buffersrc_ctx = nullptr;
+    }
+    if (buffersink_ctx) {
+        avfilter_free(buffersink_ctx);
+        buffersink_ctx = nullptr;
+    }
+    if (device_ctx) {
+        av_buffer_unref(&device_ctx);
+        device_ctx = nullptr;
+    }
+    if (filter_graph) {
+        avfilter_graph_free(&filter_graph);
+        filter_graph = nullptr;
+    }
+}
+
+int LibplaceboFilter::init(AVCodecContext *dec_ctx, AVCodecContext *enc_ctx) {
+    // Construct the shader path
+    std::filesystem::path shader_full_path;
+    if (filepath_is_readable(shader_path)) {
+        // If the shader path is directly readable, use it
+        shader_full_path = shader_path;
+    } else {
+        // Construct the fallback path using std::filesystem
+        shader_full_path =
+            find_resource_file(std::filesystem::path("models") / (shader_path.string() + ".glsl"));
+    }
+
+    // Save the output time base
+    output_time_base = enc_ctx->time_base;
+
+    return init_libplacebo(
+        &filter_graph,
+        &buffersrc_ctx,
+        &buffersink_ctx,
+        &device_ctx,
+        dec_ctx,
+        output_width,
+        output_height,
+        shader_full_path
+    );
+}
+
+AVFrame *LibplaceboFilter::process_frame(AVFrame *input_frame) {
+    int ret;
+
+    // Get the filtered frame
+    AVFrame *output_frame = av_frame_alloc();
+    if (output_frame == nullptr) {
+        fprintf(stderr, "Failed to allocate output frame\n");
+        return nullptr;
+    }
+
+    // Feed the frame to the filter graph
+    ret = av_buffersrc_add_frame(buffersrc_ctx, input_frame);
+    if (ret < 0) {
+        fprintf(stderr, "Error while feeding the filter graph\n");
+        return nullptr;
+    }
+
+    ret = av_buffersink_get_frame(buffersink_ctx, output_frame);
+    if (ret < 0) {
+        av_frame_free(&output_frame);
+        if (ret != AVERROR(EAGAIN) && ret != AVERROR_EOF) {
+            char errbuf[AV_ERROR_MAX_STRING_SIZE];
+            av_strerror(ret, errbuf, sizeof(errbuf));
+            fprintf(stderr, "Error getting frame from filter graph: %s\n", errbuf);
+            return nullptr;
+        }
+        return (AVFrame *)-1;
+    }
+
+    // Rescale PTS to encoder's time base
+    output_frame->pts =
+        av_rescale_q(output_frame->pts, buffersink_ctx->inputs[0]->time_base, output_time_base);
+
+    // Return the processed frame to the caller
+    return output_frame;
+}
+
+int LibplaceboFilter::flush(std::vector<AVFrame *> &processed_frames) {
+    int ret = av_buffersrc_add_frame(buffersrc_ctx, nullptr);  // Signal EOF to the filter graph
+    if (ret < 0) {
+        fprintf(stderr, "Error while flushing filter graph\n");
+        return ret;
+    }
+
+    // Retrieve all remaining frames from the filter graph
+    while (1) {
+        AVFrame *filt_frame = av_frame_alloc();
+        if (filt_frame == nullptr) {
+            return AVERROR(ENOMEM);
+        }
+
+        ret = av_buffersink_get_frame(buffersink_ctx, filt_frame);
+        if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {
+            av_frame_free(&filt_frame);
+            break;
+        }
+        if (ret < 0) {
+            av_frame_free(&filt_frame);
+            return ret;
+        }
+
+        // Rescale PTS to encoder's time base
+        filt_frame->pts =
+            av_rescale_q(filt_frame->pts, buffersink_ctx->inputs[0]->time_base, output_time_base);
+
+        // Add to processed frames
+        processed_frames.push_back(filt_frame);
+    }
+
+    return 0;
+}

src/libvideo2x.cpp

@@ -0,0 +1,327 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <cstdint>
+
+// FFmpeg headers
+extern "C" {
+#include <libavcodec/avcodec.h>
+#include <libavformat/avformat.h>
+}
+
+#include "decoder.h"
+#include "encoder.h"
+#include "filter.h"
+#include "libplacebo_filter.h"
+#include "libvideo2x.h"
+#include "realesrgan_filter.h"
+
+// Function to process frames using the selected filter (same as before)
+int process_frames(
+    ProcessingStatus *status,
+    AVFormatContext *fmt_ctx,
+    AVFormatContext *ofmt_ctx,
+    AVCodecContext *dec_ctx,
+    AVCodecContext *enc_ctx,
+    Filter *filter,
+    int video_stream_index
+) {
+    int ret;
+    AVPacket packet;
+    std::vector<AVFrame *> flushed_frames;
+    char errbuf[AV_ERROR_MAX_STRING_SIZE];
+
+    // Get the total number of frames in the video
+    AVStream *video_stream = fmt_ctx->streams[video_stream_index];
+    status->total_frames = video_stream->nb_frames;
+
+    // If nb_frames is not set, calculate total frames using duration and frame rate
+    if (status->total_frames == 0) {
+        int64_t duration = video_stream->duration;
+        AVRational frame_rate = video_stream->avg_frame_rate;
+        if (duration != AV_NOPTS_VALUE && frame_rate.num != 0 && frame_rate.den != 0) {
+            status->total_frames = duration * frame_rate.num / frame_rate.den;
+        }
+    }
+
+    // Get start time
+    status->start_time = time(NULL);
+    if (status->start_time == -1) {
+        perror("time");
+    }
+
+    AVFrame *frame = av_frame_alloc();
+    if (frame == nullptr) {
+        ret = AVERROR(ENOMEM);
+        goto end;
+    }
+
+    // Read frames from the input file
+    while (1) {
+        ret = av_read_frame(fmt_ctx, &packet);
+        if (ret < 0) {
+            break;  // End of file or error
+        }
+
+        if (packet.stream_index == video_stream_index) {
+            // Send the packet to the decoder
+            ret = avcodec_send_packet(dec_ctx, &packet);
+            if (ret < 0) {
+                av_strerror(ret, errbuf, sizeof(errbuf));
+                fprintf(stderr, "Error sending packet to decoder: %s\n", errbuf);
+                av_packet_unref(&packet);
+                goto end;
+            }
+
+            // Receive and process frames from the decoder
+            while (1) {
+                ret = avcodec_receive_frame(dec_ctx, frame);
+                if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {
+                    break;
+                } else if (ret < 0) {
+                    av_strerror(ret, errbuf, sizeof(errbuf));
+                    fprintf(stderr, "Error decoding video frame: %s\n", errbuf);
+                    goto end;
+                }
+
+                // Process the frame using the selected filter
+                AVFrame *processed_frame = filter->process_frame(frame);
+                if (processed_frame != nullptr && processed_frame != (AVFrame *)-1) {
+                    // Encode and write the processed frame
+                    ret = encode_and_write_frame(processed_frame, enc_ctx, ofmt_ctx);
+                    if (ret < 0) {
+                        av_strerror(ret, errbuf, sizeof(errbuf));
+                        fprintf(stderr, "Error encoding/writing frame: %s\n", errbuf);
+                        av_frame_free(&processed_frame);
+                        goto end;
+                    }
+
+                    av_frame_free(&processed_frame);
+                    status->processed_frames++;
+                } else if (processed_frame != (AVFrame *)-1) {
+                    fprintf(stderr, "Error processing frame\n");
+                    goto end;
+                }
+
+                av_frame_unref(frame);
+
+                // Print the processing status
+                printf(
+                    "\r[Video2X] Processing frame %ld/%ld (%.2f%%); time elapsed: %lds",
+                    status->processed_frames,
+                    status->total_frames,
+                    status->processed_frames * 100.0 / status->total_frames,
+                    time(NULL) - status->start_time
+                );
+                fflush(stdout);
+            }
+        }
+        av_packet_unref(&packet);
+    }
+
+    // Print a newline after processing all frames
+    printf("\n");
+
+    // Flush the filter
+    ret = filter->flush(flushed_frames);
+    if (ret < 0) {
+        av_strerror(ret, errbuf, sizeof(errbuf));
+        fprintf(stderr, "Error flushing filter: %s\n", errbuf);
+        goto end;
+    }
+
+    // Encode and write all flushed frames
+    for (AVFrame *&flushed_frame : flushed_frames) {
+        ret = encode_and_write_frame(flushed_frame, enc_ctx, ofmt_ctx);
+        if (ret < 0) {
+            av_strerror(ret, errbuf, sizeof(errbuf));
+            fprintf(stderr, "Error encoding/writing flushed frame: %s\n", errbuf);
+            av_frame_free(&flushed_frame);
+            flushed_frame = nullptr;
+            goto end;
+        }
+        av_frame_free(&flushed_frame);
+        flushed_frame = nullptr;
+    }
+
+    // Flush the encoder
+    ret = flush_encoder(enc_ctx, ofmt_ctx);
+    if (ret < 0) {
+        av_strerror(ret, errbuf, sizeof(errbuf));
+        fprintf(stderr, "Error flushing encoder: %s\n", errbuf);
+        goto end;
+    }
+
+end:
+    av_frame_free(&frame);
+    // Free any flushed frames not yet freed
+    for (AVFrame *flushed_frame : flushed_frames) {
+        if (flushed_frame) {
+            av_frame_free(&flushed_frame);
+        }
+    }
+    return ret;
+}
+
+// Cleanup helper function
+void cleanup(
+    AVFormatContext *fmt_ctx,
+    AVFormatContext *ofmt_ctx,
+    AVCodecContext *dec_ctx,
+    AVCodecContext *enc_ctx,
+    Filter *filter
+) {
+    if (filter) {
+        delete filter;
+    }
+    if (dec_ctx) {
+        avcodec_free_context(&dec_ctx);
+    }
+    if (enc_ctx) {
+        avcodec_free_context(&enc_ctx);
+    }
+    if (fmt_ctx) {
+        avformat_close_input(&fmt_ctx);
+    }
+    if (ofmt_ctx && !(ofmt_ctx->oformat->flags & AVFMT_NOFILE)) {
+        avio_closep(&ofmt_ctx->pb);
+    }
+    if (ofmt_ctx) {
+        avformat_free_context(ofmt_ctx);
+    }
+}
+
+// Main function to process the video
+extern "C" int process_video(
+    const char *input_filename,
+    const char *output_filename,
+    const FilterConfig *filter_config,
+    EncoderConfig *encoder_config,
+    ProcessingStatus *status
+) {
+    AVFormatContext *fmt_ctx = nullptr;
+    AVFormatContext *ofmt_ctx = nullptr;
+    AVCodecContext *dec_ctx = nullptr;
+    AVCodecContext *enc_ctx = nullptr;
+    Filter *filter = nullptr;
+    int video_stream_index = -1;
+    int ret = 0;  // Initialize ret with 0 to assume success
+
+    // Initialize input
+    if (init_decoder(input_filename, &fmt_ctx, &dec_ctx, &video_stream_index) < 0) {
+        fprintf(stderr, "Failed to initialize decoder\n");
+        cleanup(fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter);
+        return 1;
+    }
+
+    // Initialize output based on Libplacebo or RealESRGAN configuration
+    int output_width = 0, output_height = 0;
+    switch (filter_config->filter_type) {
+        case FILTER_LIBPLACEBO:
+            output_width = filter_config->config.libplacebo.output_width;
+            output_height = filter_config->config.libplacebo.output_height;
+            break;
+        case FILTER_REALESRGAN:
+            // Calculate the output dimensions based on the scaling factor
+            output_width = dec_ctx->width * filter_config->config.realesrgan.scaling_factor;
+            output_height = dec_ctx->height * filter_config->config.realesrgan.scaling_factor;
+    }
+
+    // Initialize output encoder
+    encoder_config->output_width = output_width;
+    encoder_config->output_height = output_height;
+    if (init_encoder(output_filename, &ofmt_ctx, &enc_ctx, dec_ctx, encoder_config) < 0) {
+        fprintf(stderr, "Failed to initialize encoder\n");
+        cleanup(fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter);
+        return 1;
+    }
+
+    // Write the output file header
+    if (avformat_write_header(ofmt_ctx, NULL) < 0) {
+        fprintf(stderr, "Error occurred when opening output file\n");
+        cleanup(fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter);
+        return 1;
+    }
+
+    // Create and initialize the appropriate filter
+    switch (filter_config->filter_type) {
+        case FILTER_LIBPLACEBO: {
+            const auto &config = filter_config->config.libplacebo;
+
+            // Validate shader path
+            if (!config.shader_path) {
+                fprintf(stderr, "Shader path must be provided for the libplacebo filter\n");
+                cleanup(fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter);
+                return 1;
+            }
+
+            // Validate output dimensions
+            if (config.output_width <= 0 || config.output_height <= 0) {
+                fprintf(stderr, "Output dimensions must be provided for the libplacebo filter\n");
+                cleanup(fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter);
+                return 1;
+            }
+
+            filter = new LibplaceboFilter(
+                config.output_width, config.output_height, std::filesystem::path(config.shader_path)
+            );
+            break;
+        }
+        case FILTER_REALESRGAN: {
+            const auto &config = filter_config->config.realesrgan;
+
+            // Validate model name
+            if (!config.model) {
+                fprintf(stderr, "Model name must be provided for the RealESRGAN filter\n");
+                cleanup(fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter);
+                return 1;
+            }
+
+            // Validate scaling factor
+            if (config.scaling_factor <= 0) {
+                fprintf(stderr, "Scaling factor must be provided for the RealESRGAN filter\n");
+                cleanup(fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter);
+                return 1;
+            }
+
+            filter = new RealesrganFilter(
+                config.gpuid, config.tta_mode, config.scaling_factor, config.model
+            );
+            break;
+        }
+        default:
+            fprintf(stderr, "Unknown filter type\n");
+            cleanup(fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter);
+            return 1;
+    }
+
+    // Initialize the filter
+    if (filter->init(dec_ctx, enc_ctx) < 0) {
+        fprintf(stderr, "Failed to initialize filter\n");
+        cleanup(fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter);
+        return 1;
+    }
+
+    // Process frames
+    if ((ret =
+             process_frames(status, fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter, video_stream_index)
+        ) < 0) {
+        fprintf(stderr, "Error processing frames\n");
+        cleanup(fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter);
+        return 1;
+    }
+
+    // Write the output file trailer
+    av_write_trailer(ofmt_ctx);
+
+    // Cleanup before returning
+    cleanup(fmt_ctx, ofmt_ctx, dec_ctx, enc_ctx, filter);
+
+    if (ret < 0 && ret != AVERROR_EOF) {
+        char errbuf[AV_ERROR_MAX_STRING_SIZE];
+        av_strerror(ret, errbuf, sizeof(errbuf));
+        fprintf(stderr, "Error occurred: %s\n", errbuf);
+        return 1;
+    }
+    return 0;
+}

src/locale/zh_CN/LC_MESSAGES/zh_CN.po

@@ -1,298 +0,0 @@
-# SOME DESCRIPTIVE TITLE.
-# Copyright (C) YEAR ORGANIZATION
-# FIRST AUTHOR <EMAIL@ADDRESS>, YEAR.
-#
-msgid ""
-msgstr ""
-"Project-Id-Version: \n"
-"POT-Creation-Date: 2020-05-11 04:39-0400\n"
-"PO-Revision-Date: 2020-05-11 04:40-0400\n"
-"Last-Translator: \n"
-"Language-Team: \n"
-"Language: zh_CN\n"
-"MIME-Version: 1.0\n"
-"Content-Type: text/plain; charset=UTF-8\n"
-"Content-Transfer-Encoding: 8bit\n"
-"Generated-By: pygettext.py 1.5\n"
-"X-Generator: Poedit 2.3.1\n"
-"Plural-Forms: nplurals=1; plural=0;\n"
-
-#: progress_monitor.py:42
-msgid "Upscaling Progress"
-msgstr "放大进度"
-
-#: upscaler.py:104
-msgid "Specified or default cache directory is a file/link"
-msgstr "指定或默认的缓存目录是文件/链接"
-
-#: upscaler.py:110
-msgid "Creating cache directory {}"
-msgstr "创建缓存目录 {}"
-
-#: upscaler.py:113
-msgid "Unable to create {}"
-msgstr "无法创建 {}"
-
-#: upscaler.py:118
-msgid "Extracted frames are being saved to: {}"
-msgstr "提取的帧将被保存到：{}"
-
-#: upscaler.py:120
-msgid "Upscaled frames are being saved to: {}"
-msgstr "已放大的帧将被保存到：{}"
-
-#: upscaler.py:130
-msgid "Cleaning up cache directory: {}"
-msgstr "清理缓存目录：{}"
-
-#: upscaler.py:133
-msgid "Unable to delete: {}"
-msgstr "无法删除：{}"
-
-#: upscaler.py:139 upscaler.py:154 upscaler.py:165
-msgid "Input and output path type mismatch"
-msgstr "输入和输出路径类型不匹配"
-
-#: upscaler.py:140
-msgid "Input is multiple files but output is not directory"
-msgstr "输入是多个文件，但输出不是目录"
-
-#: upscaler.py:144
-msgid "Input path {} is neither a file nor a directory"
-msgstr "输入路径 {} 既不是文件也不是目录"
-
-#: upscaler.py:148 upscaler.py:170
-msgid "Input directory and output directory cannot be the same"
-msgstr "输入目录和输出目录不能相同"
-
-#: upscaler.py:155
-msgid "Input is single file but output is directory"
-msgstr "所选的输入路径是单个文件，但输出路径是目录"
-
-#: upscaler.py:158
-msgid "No suffix found in output file path"
-msgstr "在输出文件路径中未找到后缀"
-
-#: upscaler.py:159
-msgid "Suffix must be specified for FFmpeg"
-msgstr "必须为 FFmpeg 指定后缀"
-
-#: upscaler.py:166
-msgid "Input is directory but output is existing single file"
-msgstr "输入是目录，但输出是现有的单个文件"
-
-#: upscaler.py:175
-msgid "Input path is neither a file nor a directory"
-msgstr "输入路径既不是文件也不是目录"
-
-#: upscaler.py:184
-msgid "FFmpeg or FFprobe cannot be found under the specified path"
-msgstr "在指定的路径下找不到 FFmpeg 或 FFprobe"
-
-#: upscaler.py:185 upscaler.py:195
-msgid "Please check the configuration file settings"
-msgstr "请检查配置文件设置"
-
-#: upscaler.py:194
-msgid "Specified driver executable directory doesn't exist"
-msgstr "指定驱动的可执行文件不存在"
-
-#: upscaler.py:221
-msgid "Failed to parse driver argument: {}"
-msgstr "解析驱动程序参数失败：{}"
-
-#: upscaler.py:253
-msgid "Unrecognized driver: {}"
-msgstr "无法识别的驱动名称：{}"
-
-#: upscaler.py:293
-msgid "Starting progress monitor"
-msgstr "启动进度监视器"
-
-#: upscaler.py:298
-msgid "Starting upscaled image cleaner"
-msgstr "启动已放大图像清理程序"
-
-#: upscaler.py:307 upscaler.py:324
-msgid "Killing progress monitor"
-msgstr "终结进度监视器"
-
-#: upscaler.py:310 upscaler.py:327
-msgid "Killing upscaled image cleaner"
-msgstr "终结已放大图像清理程序"
-
-#: upscaler.py:331
-msgid "Terminating all processes"
-msgstr "正在终止所有进程"
-
-#: upscaler.py:338
-msgid "Main process waiting for subprocesses to exit"
-msgstr "主进程开始等待子进程结束"
-
-#: upscaler.py:357 upscaler.py:361
-msgid "Subprocess {} exited with code {}"
-msgstr "子进程 {} 结束，返回码 {}"
-
-#: upscaler.py:367
-msgid "Stop signal received"
-msgstr "收到停止信号"
-
-#: upscaler.py:372
-msgid "Subprocess execution ran into an error"
-msgstr "子进程执行遇到错误"
-
-#: upscaler.py:421
-msgid "Upscaling single video file: {}"
-msgstr "放大单个视频文件：{}"
-
-#: upscaler.py:443 upscaler.py:502
-msgid "Starting to upscale extracted images"
-msgstr "开始对提取的帧进行放大"
-
-#: upscaler.py:448 upscaler.py:504
-msgid "Upscaling completed"
-msgstr "放大完成"
-
-#: upscaler.py:457
-msgid "Reading video information"
-msgstr "读取视频信息"
-
-#: upscaler.py:471
-msgid "Aborting: No video stream found"
-msgstr "程序中止：文件中未找到视频流"
-
-#: upscaler.py:490
-msgid "Unsupported pixel format: {}"
-msgstr "不支持的像素格式：{}"
-
-#: upscaler.py:493
-msgid "Framerate: {}"
-msgstr "帧率：{}"
-
-#: upscaler.py:507
-msgid "Converting extracted frames into video"
-msgstr "正在将提取的帧转换为视频"
-
-#: upscaler.py:514
-msgid "Conversion completed"
-msgstr "转换已完成"
-
-#: upscaler.py:518
-msgid "Migrating audio, subtitles and other streams to upscaled video"
-msgstr "正在将音频、字幕和其他流迁移到放大后的视频"
-
-#: upscaler.py:527
-msgid "Failed to migrate streams"
-msgstr "迁移流失败"
-
-#: upscaler.py:528
-msgid "Trying to output video without additional streams"
-msgstr "正在尝试输出不含其他流的视频"
-
-#: upscaler.py:535
-msgid "Output video file exists, aborting"
-msgstr "输出目标文件已存在，取消输出"
-
-#: upscaler.py:539
-msgid "Writing intermediate file to: {}"
-msgstr "正在将中间视频文件写入至：{}"
-
-#: video2x.py:84
-msgid ""
-"Video2X Version: {}\n"
-"Author: K4YT3X\n"
-"License: GNU GPL v3\n"
-"Github Page: https://github.com/k4yt3x/video2x\n"
-"Contact: k4yt3x@k4yt3x.com"
-msgstr ""
-"Video2X 版本： {}\n"
-"作者： K4YT3X\n"
-"开源许可： GNU GPL v3\n"
-"GitHub 主页：https://github.com/k4yt3x/video2x\n"
-"联系方式：k4yt3x@k4yt3x.com"
-
-#: video2x.py:106
-msgid "Video2X Options"
-msgstr "Video2X 选项"
-
-#: video2x.py:107
-msgid "show this help message and exit"
-msgstr "显示此帮助消息并退出"
-
-#: video2x.py:108
-msgid "source video file/directory"
-msgstr "源视频文件/目录"
-
-#: video2x.py:109
-msgid "output video file/directory"
-msgstr "输出视频文件/目录"
-
-#: video2x.py:110
-msgid "video2x config file path"
-msgstr "video2x 配置文件路径"
-
-#: video2x.py:112
-msgid "display version, lawful information and exit"
-msgstr "显示版本和法律信息并退出"
-
-#: video2x.py:115
-msgid "Upscaling Options"
-msgstr "视频放大选项"
-
-#: video2x.py:116
-msgid "upscaling driver"
-msgstr "视频放大驱动"
-
-#: video2x.py:117
-msgid "scaling ratio"
-msgstr "缩放比"
-
-#: video2x.py:118
-msgid "number of processes to use for upscaling"
-msgstr "并发进程数"
-
-#: video2x.py:119
-msgid "preserve extracted and upscaled frames"
-msgstr "保留提取的和放大的帧"
-
-#: video2x.py:159
-msgid "This file cannot be imported"
-msgstr "此文件无法被当作模块导入"
-
-#: video2x.py:229
-msgid "Program completed, taking {} seconds"
-msgstr "程序执行完毕，总计花费 {} 秒"
-
-#: video2x.py:232
-msgid "An exception has occurred"
-msgstr "发生了异常"
-
-#~ msgid "output video width"
-#~ msgstr "输出视频宽度"
-
-#~ msgid "output video height"
-#~ msgstr "输出视频高度"
-
-#~ msgid "You must specify input video file/directory path"
-#~ msgstr "您必须指定输入视频文件/目录路径"
-
-#~ msgid "You must specify output video file/directory path"
-#~ msgstr "您必须指定输出视频文件/目录路径"
-
-#~ msgid "Selected driver accepts only scaling ratio"
-#~ msgstr "所选驱动程序仅接受缩放比率"
-
-#~ msgid "Scaling ratio must be 1 or 2 for waifu2x_ncnn_vulkan"
-#~ msgstr "waifu2x_ncnn_vulkan 的缩放比必须为 1 或 2"
-
-#~ msgid "Scaling ratio must be one of 2, 3 or 4 for srmd_ncnn_vulkan"
-#~ msgstr "srmd_ncnn_vulkan 的缩放比必须为 2、3 或 4"
-
-#~ msgid "You can only specify either scaling ratio or output width and height"
-#~ msgstr "您只能指定缩放比或输出宽度和高度两者之一"
-
-#~ msgid "You must specify both width and height"
-#~ msgstr "您必须同时指定宽度和高度"
-
-#~ msgid "Upscaling videos in directory: {}"
-#~ msgstr "放大该文件夹中的所有视频：{}"

src/progress_monitor.py

@@ -1,64 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Name: Video2X Upscale Progress Monitor
-Author: BrianPetkovsek
-Date Created: May 7, 2020
-Last Modified: May 10, 2020
-"""
-
-# built-in imports
-import contextlib
-import threading
-import time
-
-# third-party imports
-from tqdm import tqdm
-
-
-class ProgressMonitor(threading.Thread):
-    """ progress monitor
-
-    This class provides progress monitoring functionalities
-    by keeping track of the amount of frames in the input
-    directory and the output directory. This is originally
-    suggested by @ArmandBernard.
-    """
-
-    def __init__(self, upscaler, extracted_frames_directories):
-        threading.Thread.__init__(self)
-        self.upscaler = upscaler
-        self.extracted_frames_directories = extracted_frames_directories
-        self.running = False
-
-    def run(self):
-        self.running = True
-
-        # get number of extracted frames
-        self.upscaler.total_frames = 0
-        for directory in self.extracted_frames_directories:
-            self.upscaler.total_frames += len([f for f in directory.iterdir() if str(f).lower().endswith(self.upscaler.image_format.lower())])
-
-        with tqdm(total=self.upscaler.total_frames, ascii=True, desc=_('Upscaling Progress')) as progress_bar:
-            # tqdm update method adds the value to the progress
-            # bar instead of setting the value. Therefore, a delta
-            # needs to be calculated.
-            previous_cycle_frames = 0
-            while self.running:
-
-                with contextlib.suppress(FileNotFoundError):
-                    upscaled_frames = [f for f in self.upscaler.upscaled_frames.iterdir() if str(f).lower().endswith(self.upscaler.image_format.lower())]
-                    if len(upscaled_frames) >= 1:
-                        self.upscaler.last_frame_upscaled = sorted(upscaled_frames)[-1]
-                    self.upscaler.total_frames_upscaled = len(upscaled_frames)
-
-                    # update progress bar
-                    delta = self.upscaler.total_frames_upscaled - previous_cycle_frames
-                    previous_cycle_frames = self.upscaler.total_frames_upscaled
-                    progress_bar.update(delta)
-
-                time.sleep(1)
-
-    def stop(self):
-        self.running = False
-        self.join()

src/realesrgan_filter.cpp

@@ -0,0 +1,129 @@
+#include <cstdint>
+#include <cstdio>
+#include <filesystem>
+#include <string>
+
+extern "C" {
+#include <libavcodec/avcodec.h>
+#include <libavutil/avutil.h>
+#include <libavutil/imgutils.h>
+}
+
+#include "conversions.h"
+#include "fsutils.h"
+#include "realesrgan.h"
+#include "realesrgan_filter.h"
+
+RealesrganFilter::RealesrganFilter(
+    int gpuid,
+    bool tta_mode,
+    int scaling_factor,
+    const char *model,
+    const std::filesystem::path custom_model_param_path,
+    const std::filesystem::path custom_model_bin_path
+)
+    : realesrgan(nullptr),
+      gpuid(gpuid),
+      tta_mode(tta_mode),
+      scaling_factor(scaling_factor),
+      model(model),
+      custom_model_param_path(std::move(custom_model_param_path)),
+      custom_model_bin_path(std::move(custom_model_bin_path)) {}
+
+RealesrganFilter::~RealesrganFilter() {
+    if (realesrgan) {
+        delete realesrgan;
+        realesrgan = nullptr;
+    }
+}
+
+int RealesrganFilter::init(AVCodecContext *dec_ctx, AVCodecContext *enc_ctx) {
+    // Construct the model paths using std::filesystem
+    std::filesystem::path model_param_path;
+    std::filesystem::path model_bin_path;
+
+    if (model) {
+        // Find the model paths by model name if provided
+        model_param_path = std::filesystem::path("models") /
+                           (std::string(model) + "-x" + std::to_string(scaling_factor) + ".param");
+        model_bin_path = std::filesystem::path("models") /
+                         (std::string(model) + "-x" + std::to_string(scaling_factor) + ".bin");
+    } else if (!custom_model_param_path.empty() && !custom_model_bin_path.empty()) {
+        // Use the custom model paths if provided
+        model_param_path = custom_model_param_path;
+        model_bin_path = custom_model_bin_path;
+    } else {
+        // Neither model name nor custom model paths provided
+        fprintf(stderr, "Model or model paths must be provided for RealESRGAN filter\n");
+        return -1;
+    }
+
+    // Get the full paths using a function that possibly modifies or validates the path
+    std::filesystem::path model_param_full_path = find_resource_file(model_param_path);
+    std::filesystem::path model_bin_full_path = find_resource_file(model_bin_path);
+
+    // Create a new RealESRGAN instance
+    realesrgan = new RealESRGAN(gpuid, tta_mode);
+
+    // Store the time bases
+    input_time_base = dec_ctx->time_base;
+    output_time_base = enc_ctx->time_base;
+    output_pix_fmt = enc_ctx->pix_fmt;
+
+    // Load the model
+    if (realesrgan->load(model_param_full_path, model_bin_full_path) != 0) {
+        fprintf(stderr, "Failed to load RealESRGAN model\n");
+        return -1;
+    }
+
+    // Set RealESRGAN parameters
+    realesrgan->scale = scaling_factor;
+    realesrgan->prepadding = 10;
+
+    // Calculate tilesize based on GPU heap budget
+    uint32_t heap_budget = ncnn::get_gpu_device(gpuid)->get_heap_budget();
+    if (heap_budget > 1900) {
+        realesrgan->tilesize = 200;
+    } else if (heap_budget > 550) {
+        realesrgan->tilesize = 100;
+    } else if (heap_budget > 190) {
+        realesrgan->tilesize = 64;
+    } else {
+        realesrgan->tilesize = 32;
+    }
+
+    return 0;
+}
+
+AVFrame *RealesrganFilter::process_frame(AVFrame *input_frame) {
+    // Convert the input frame to RGB24
+    ncnn::Mat input_mat = avframe_to_ncnn_mat(input_frame);
+    if (input_mat.empty()) {
+        fprintf(stderr, "Failed to convert AVFrame to ncnn::Mat\n");
+        return nullptr;
+    }
+
+    // Allocate space for ouptut ncnn::Mat
+    int output_width = input_mat.w * realesrgan->scale;
+    int output_height = input_mat.h * realesrgan->scale;
+    ncnn::Mat output_mat = ncnn::Mat(output_width, output_height, (size_t)3, 3);
+
+    if (realesrgan->process(input_mat, output_mat) != 0) {
+        fprintf(stderr, "RealESRGAN processing failed\n");
+        return nullptr;
+    }
+
+    // Convert ncnn::Mat to AVFrame
+    AVFrame *output_frame = ncnn_mat_to_avframe(output_mat, output_pix_fmt);
+
+    // Rescale PTS to encoder's time base
+    output_frame->pts = av_rescale_q(input_frame->pts, input_time_base, output_time_base);
+
+    // Return the processed frame to the caller
+    return output_frame;
+}
+
+int RealesrganFilter::flush(std::vector<AVFrame *> &processed_frames) {
+    // No special flushing needed for RealESRGAN
+    return 0;
+}

src/requirements.txt

@@ -1,9 +0,0 @@
-avalon_framework
-colorama
-patool
-psutil
-pyqt5
-pyunpack
-pyyaml
-requests
-tqdm

src/upscaler.py

@@ -1,555 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Name: Video2X Upscaler
-Author: K4YT3X
-Date Created: December 10, 2018
-Last Modified: May 10, 2020
-
-Description: This file contains the Upscaler class. Each
-instance of the Upscaler class is an upscaler on an image or
-a folder.
-"""
-
-# local imports
-from exceptions import *
-from image_cleaner import ImageCleaner
-from progress_monitor import ProgressMonitor
-from wrappers.ffmpeg import Ffmpeg
-
-# built-in imports
-from fractions import Fraction
-import contextlib
-import copy
-import gettext
-import importlib
-import locale
-import os
-import pathlib
-import queue
-import re
-import shutil
-import subprocess
-import tempfile
-import time
-import traceback
-
-# third-party imports
-from avalon_framework import Avalon
-
-# internationalization constants
-DOMAIN = 'video2x'
-LOCALE_DIRECTORY = pathlib.Path(__file__).parent.absolute() / 'locale'
-
-# getting default locale settings
-default_locale, encoding = locale.getdefaultlocale()
-language = gettext.translation(DOMAIN, LOCALE_DIRECTORY, [default_locale], fallback=True)
-language.install()
-_ = language.gettext
-
-# these names are consistent for
-# - driver selection in command line
-# - driver wrapper file names
-# - config file keys
-AVAILABLE_DRIVERS = ['waifu2x_caffe',
-                     'waifu2x_converter_cpp',
-                     'waifu2x_ncnn_vulkan',
-                     'srmd_ncnn_vulkan',
-                     'anime4kcpp']
-
-
-class Upscaler:
-    """ An instance of this class is a upscaler that will
-    upscale all images in the given directory.
-
-    Raises:
-        Exception -- all exceptions
-        ArgumentError -- if argument is not valid
-    """
-
-    def __init__(self, input_path, output_path, driver_settings, ffmpeg_settings):
-        # mandatory arguments
-        self.input = input_path
-        self.output = output_path
-        self.driver_settings = driver_settings
-        self.ffmpeg_settings = ffmpeg_settings
-
-        # optional arguments
-        self.driver = 'waifu2x_caffe'
-        self.scale_ratio = None
-        self.processes = 1
-        self.video2x_cache_directory = pathlib.Path(tempfile.gettempdir()) / 'video2x'
-        self.image_format = 'png'
-        self.preserve_frames = False
-
-        # other internal members and signals
-        self.running = False
-        self.total_frames_upscaled = 0
-        self.total_frames = 0
-        self.total_videos = 0
-        self.total_processed = 0
-        self.current_input_video = pathlib.Path()
-        self.last_frame_upscaled = pathlib.Path()
-
-    def create_temp_directories(self):
-        """create temporary directories
-        """
-
-        # if cache directory unspecified, use %TEMP%\video2x
-        if self.video2x_cache_directory is None:
-            self.video2x_cache_directory = pathlib.Path(tempfile.gettempdir()) / 'video2x'
-
-        # if specified cache path exists and isn't a directory
-        if self.video2x_cache_directory.exists() and not self.video2x_cache_directory.is_dir():
-            Avalon.error(_('Specified or default cache directory is a file/link'))
-            raise FileExistsError('Specified or default cache directory is a file/link')
-
-        # if cache directory doesn't exist, try creating it
-        if not self.video2x_cache_directory.exists():
-            try:
-                Avalon.debug_info(_('Creating cache directory {}').format(self.video2x_cache_directory))
-                self.video2x_cache_directory.mkdir(parents=True, exist_ok=True)
-            except Exception as exception:
-                Avalon.error(_('Unable to create {}').format(self.video2x_cache_directory))
-                raise exception
-
-        # create temp directories for extracted frames and upscaled frames
-        self.extracted_frames = pathlib.Path(tempfile.mkdtemp(dir=self.video2x_cache_directory))
-        Avalon.debug_info(_('Extracted frames are being saved to: {}').format(self.extracted_frames))
-        self.upscaled_frames = pathlib.Path(tempfile.mkdtemp(dir=self.video2x_cache_directory))
-        Avalon.debug_info(_('Upscaled frames are being saved to: {}').format(self.upscaled_frames))
-
-    def cleanup_temp_directories(self):
-        """delete temp directories when done
-        """
-        if not self.preserve_frames:
-            for directory in [self.extracted_frames, self.upscaled_frames, self.video2x_cache_directory]:
-                try:
-                    # avalon framework cannot be used if python is shutting down
-                    # therefore, plain print is used
-                    print(_('Cleaning up cache directory: {}').format(directory))
-                    shutil.rmtree(directory)
-                except (OSError, FileNotFoundError):
-                    print(_('Unable to delete: {}').format(directory))
-                    traceback.print_exc()
-
-    def _check_arguments(self):
-        if isinstance(self.input, list):
-            if self.output.exists() and not self.output.is_dir():
-                Avalon.error(_('Input and output path type mismatch'))
-                Avalon.error(_('Input is multiple files but output is not directory'))
-                raise ArgumentError('input output path type mismatch')
-            for input_path in self.input:
-                if not input_path.is_file() and not input_path.is_dir():
-                    Avalon.error(_('Input path {} is neither a file nor a directory').format(input_path))
-                    raise FileNotFoundError(f'{input_path} is neither file nor directory')
-                with contextlib.suppress(FileNotFoundError):
-                    if input_path.samefile(self.output):
-                        Avalon.error(_('Input directory and output directory cannot be the same'))
-                        raise FileExistsError('input directory and output directory are the same')
-
-        # if input is a file
-        elif self.input.is_file():
-            if self.output.is_dir():
-                Avalon.error(_('Input and output path type mismatch'))
-                Avalon.error(_('Input is single file but output is directory'))
-                raise ArgumentError('input output path type mismatch')
-            if not re.search(r'.*\..*$', str(self.output)):
-                Avalon.error(_('No suffix found in output file path'))
-                Avalon.error(_('Suffix must be specified for FFmpeg'))
-                raise ArgumentError('no output video suffix specified')
-
-        # if input is a directory
-        elif self.input.is_dir():
-            if self.output.is_file():
-                Avalon.error(_('Input and output path type mismatch'))
-                Avalon.error(_('Input is directory but output is existing single file'))
-                raise ArgumentError('input output path type mismatch')
-            with contextlib.suppress(FileNotFoundError):
-                if self.input.samefile(self.output):
-                    Avalon.error(_('Input directory and output directory cannot be the same'))
-                    raise FileExistsError('input directory and output directory are the same')
-
-        # if input is neither
-        else:
-            Avalon.error(_('Input path is neither a file nor a directory'))
-            raise FileNotFoundError(f'{self.input} is neither file nor directory')
-
-        # check FFmpeg settings
-        ffmpeg_path = pathlib.Path(self.ffmpeg_settings['ffmpeg_path'])
-        if not ((pathlib.Path(ffmpeg_path / 'ffmpeg.exe').is_file() and
-                pathlib.Path(ffmpeg_path / 'ffprobe.exe').is_file()) or
-                (pathlib.Path(ffmpeg_path / 'ffmpeg').is_file() and
-                pathlib.Path(ffmpeg_path / 'ffprobe').is_file())):
-            Avalon.error(_('FFmpeg or FFprobe cannot be found under the specified path'))
-            Avalon.error(_('Please check the configuration file settings'))
-            raise FileNotFoundError(self.ffmpeg_settings['ffmpeg_path'])
-
-        # check if driver settings
-        driver_settings = copy.deepcopy(self.driver_settings)
-        driver_path = driver_settings.pop('path')
-
-        # check if driver path exists
-        if not (pathlib.Path(driver_path).is_file() or pathlib.Path(f'{driver_path}.exe').is_file()):
-            Avalon.error(_('Specified driver executable directory doesn\'t exist'))
-            Avalon.error(_('Please check the configuration file settings'))
-            raise FileNotFoundError(driver_path)
-
-        # parse driver arguments using driver's parser
-        # the parser will throw AttributeError if argument doesn't satisfy constraints
-        try:
-            driver_arguments = []
-            for key in driver_settings.keys():
-
-                value = driver_settings[key]
-
-                if value is None or value is False:
-                    continue
-
-                else:
-                    if len(key) == 1:
-                        driver_arguments.append(f'-{key}')
-                    else:
-                        driver_arguments.append(f'--{key}')
-                    # true means key is an option
-                    if value is not True:
-                        driver_arguments.append(str(value))
-
-            DriverWrapperMain = getattr(importlib.import_module(f'wrappers.{self.driver}'), 'WrapperMain')
-            DriverWrapperMain.parse_arguments(driver_arguments)
-        except AttributeError as e:
-            Avalon.error(_('Failed to parse driver argument: {}').format(e.args[0]))
-            raise e
-
-        # waifu2x-caffe scale_ratio, scale_width and scale_height check
-        if self.driver == 'waifu2x_caffe':
-            if (driver_settings['scale_width'] != 0 and driver_settings['scale_height'] == 0 or
-                    driver_settings['scale_width'] == 0 and driver_settings['scale_height'] != 0):
-                Avalon.error('Only one of scale_width and scale_height is specified for waifu2x-caffe')
-                raise AttributeError('only one of scale_width and scale_height is specified for waifu2x-caffe')
-
-            # if scale_width and scale_height are specified, ensure scale_ratio is None
-            elif self.driver_settings['scale_width'] != 0 and self.driver_settings['scale_height'] != 0:
-                self.driver_settings['scale_ratio'] = None
-
-            # if scale_width and scale_height not specified
-            # ensure they are None, not 0
-            else:
-                self.driver_settings['scale_width'] = None
-                self.driver_settings['scale_height'] = None
-
-    def _upscale_frames(self):
-        """ Upscale video frames with waifu2x-caffe
-
-        This function upscales all the frames extracted
-        by ffmpeg using the waifu2x-caffe binary.
-
-        Arguments:
-            w2 {Waifu2x Object} -- initialized waifu2x object
-        """
-
-        # initialize waifu2x driver
-        if self.driver not in AVAILABLE_DRIVERS:
-            raise UnrecognizedDriverError(_('Unrecognized driver: {}').format(self.driver))
-
-        # list all images in the extracted frames
-        frames = [(self.extracted_frames / f) for f in self.extracted_frames.iterdir() if f.is_file]
-
-        # if we have less images than processes,
-        # create only the processes necessary
-        if len(frames) < self.processes:
-            self.processes = len(frames)
-
-        # create a directory for each process and append directory
-        # name into a list
-        process_directories = []
-        for process_id in range(self.processes):
-            process_directory = self.extracted_frames / str(process_id)
-            process_directories.append(process_directory)
-
-            # delete old directories and create new directories
-            if process_directory.is_dir():
-                shutil.rmtree(process_directory)
-            process_directory.mkdir(parents=True, exist_ok=True)
-
-        # waifu2x-converter-cpp will perform multi-threading within its own process
-        if self.driver in ['waifu2x_converter_cpp', 'waifu2x_ncnn_vulkan', 'srmd_ncnn_vulkan']:
-            process_directories = [self.extracted_frames]
-
-        else:
-            # evenly distribute images into each directory
-            # until there is none left in the directory
-            for image in frames:
-                # move image
-                image.rename(process_directories[0] / image.name)
-                # rotate list
-                process_directories = process_directories[-1:] + process_directories[:-1]
-
-        # create driver processes and start them
-        for process_directory in process_directories:
-            self.process_pool.append(self.driver_object.upscale(process_directory, self.upscaled_frames))
-
-        # start progress bar in a different thread
-        Avalon.debug_info(_('Starting progress monitor'))
-        self.progress_monitor = ProgressMonitor(self, process_directories)
-        self.progress_monitor.start()
-
-        # create the clearer and start it
-        Avalon.debug_info(_('Starting upscaled image cleaner'))
-        self.image_cleaner = ImageCleaner(self.extracted_frames, self.upscaled_frames, len(self.process_pool))
-        self.image_cleaner.start()
-
-        # wait for all process to exit
-        try:
-            self._wait()
-        except (Exception, KeyboardInterrupt, SystemExit) as e:
-            # cleanup
-            Avalon.debug_info(_('Killing progress monitor'))
-            self.progress_monitor.stop()
-
-            Avalon.debug_info(_('Killing upscaled image cleaner'))
-            self.image_cleaner.stop()
-            raise e
-
-        # if the driver is waifu2x-converter-cpp
-        # images need to be renamed to be recognizable for FFmpeg
-        if self.driver == 'waifu2x_converter_cpp':
-            for image in [f for f in self.upscaled_frames.iterdir() if f.is_file()]:
-                renamed = re.sub(f'_\\[.*\\]\\[x(\\d+(\\.\\d+)?)\\]\\.{self.image_format}',
-                                 f'.{self.image_format}',
-                                 str(image.name))
-                (self.upscaled_frames / image).rename(self.upscaled_frames / renamed)
-
-        # upscaling done, kill helper threads
-        Avalon.debug_info(_('Killing progress monitor'))
-        self.progress_monitor.stop()
-
-        Avalon.debug_info(_('Killing upscaled image cleaner'))
-        self.image_cleaner.stop()
-
-    def _terminate_subprocesses(self):
-        Avalon.warning(_('Terminating all processes'))
-        for process in self.process_pool:
-            process.terminate()
-
-    def _wait(self):
-        """ wait for subprocesses in process pool to complete
-        """
-        Avalon.debug_info(_('Main process waiting for subprocesses to exit'))
-
-        try:
-            # while process pool not empty
-            while self.process_pool:
-
-                # if stop signal received, terminate all processes
-                if self.running is False:
-                    raise SystemExit
-
-                for process in self.process_pool:
-                    process_status = process.poll()
-
-                    # if process finished
-                    if process_status is None:
-                        continue
-
-                    # if return code is not 0
-                    elif process_status != 0:
-                        Avalon.error(_('Subprocess {} exited with code {}').format(process.pid, process_status))
-                        raise subprocess.CalledProcessError(process_status, process.args)
-
-                    else:
-                        Avalon.debug_info(_('Subprocess {} exited with code {}').format(process.pid, process_status))
-                        self.process_pool.remove(process)
-
-                time.sleep(0.1)
-
-        except (KeyboardInterrupt, SystemExit) as e:
-            Avalon.warning(_('Stop signal received'))
-            self._terminate_subprocesses()
-            raise e
-
-        except (Exception, subprocess.CalledProcessError) as e:
-            Avalon.error(_('Subprocess execution ran into an error'))
-            self._terminate_subprocesses()
-            raise e
-
-    def run(self):
-        """ Main controller for Video2X
-
-        This function controls the flow of video conversion
-        and handles all necessary functions.
-        """
-
-        # external stop signal when called in a thread
-        self.running = True
-
-        # define process pool to contain processes
-        self.process_pool = []
-
-        # load driver modules
-        DriverWrapperMain = getattr(importlib.import_module(f'wrappers.{self.driver}'), 'WrapperMain')
-        self.driver_object = DriverWrapperMain(self.driver_settings)
-
-        # load options from upscaler class into driver settings
-        self.driver_object.load_configurations(self)
-
-        # parse arguments for waifu2x
-        # check argument sanity
-        self._check_arguments()
-
-        # define processing queue
-        self.processing_queue = queue.Queue()
-
-        # if input is a list of files
-        if isinstance(self.input, list):
-            # make output directory if it doesn't exist
-            self.output.mkdir(parents=True, exist_ok=True)
-
-            for input_path in self.input:
-
-                if input_path.is_file():
-                    output_video = self.output / input_path.name
-                    self.processing_queue.put((input_path.absolute(), output_video.absolute()))
-
-                elif input_path.is_dir():
-                    for input_video in [f for f in input_path.iterdir() if f.is_file()]:
-                        output_video = self.output / input_video.name
-                        self.processing_queue.put((input_video.absolute(), output_video.absolute()))
-
-        # if input specified is single file
-        elif self.input.is_file():
-            Avalon.info(_('Upscaling single video file: {}').format(self.input))
-            self.processing_queue.put((self.input.absolute(), self.output.absolute()))
-
-        # if input specified is a directory
-        elif self.input.is_dir():
-
-            # make output directory if it doesn't exist
-            self.output.mkdir(parents=True, exist_ok=True)
-            for input_video in [f for f in self.input.iterdir() if f.is_file()]:
-                output_video = self.output / input_video.name
-                self.processing_queue.put((input_video.absolute(), output_video.absolute()))
-
-        # record video count for external calls
-        self.total_videos = self.processing_queue.qsize()
-
-        while not self.processing_queue.empty():
-            self.current_input_video, output_video = self.processing_queue.get()
-            # drivers that have native support for video processing
-            if self.driver == 'anime4kcpp':
-                # append FFmpeg path to the end of PATH
-                # Anime4KCPP will then use FFmpeg to migrate audio tracks
-                os.environ['PATH'] += f';{self.ffmpeg_settings["ffmpeg_path"]}'
-                Avalon.info(_('Starting to upscale extracted images'))
-
-                # run Anime4KCPP
-                self.process_pool.append(self.driver_object.upscale(self.current_input_video, output_video))
-                self._wait()
-                Avalon.info(_('Upscaling completed'))
-
-            else:
-                try:
-                    self.create_temp_directories()
-
-                    # initialize objects for ffmpeg and waifu2x-caffe
-                    fm = Ffmpeg(self.ffmpeg_settings, self.image_format)
-
-                    Avalon.info(_('Reading video information'))
-                    video_info = fm.get_video_info(self.current_input_video)
-                    # analyze original video with FFprobe and retrieve framerate
-                    # width, height = info['streams'][0]['width'], info['streams'][0]['height']
-
-                    # find index of video stream
-                    video_stream_index = None
-                    for stream in video_info['streams']:
-                        if stream['codec_type'] == 'video':
-                            video_stream_index = stream['index']
-                            break
-
-                    # exit if no video stream found
-                    if video_stream_index is None:
-                        Avalon.error(_('Aborting: No video stream found'))
-                        raise StreamNotFoundError('no video stream found')
-
-                    # extract frames from video
-                    self.process_pool.append((fm.extract_frames(self.current_input_video, self.extracted_frames)))
-                    self._wait()
-
-                    # get average frame rate of video stream
-                    framerate = float(Fraction(video_info['streams'][video_stream_index]['r_frame_rate']))
-                    fm.pixel_format = video_info['streams'][video_stream_index]['pix_fmt']
-
-                    if self.driver == 'waifu2x_caffe':
-                        # get a dict of all pixel formats and corresponding bit depth
-                        pixel_formats = fm.get_pixel_formats()
-
-                        # try getting pixel format's corresponding bti depth
-                        try:
-                            self.driver_settings['output_depth'] = pixel_formats[fm.pixel_format]
-                        except KeyError:
-                            Avalon.error(_('Unsupported pixel format: {}').format(fm.pixel_format))
-                            raise UnsupportedPixelError(f'unsupported pixel format {fm.pixel_format}')
-
-                    Avalon.info(_('Framerate: {}').format(framerate))
-
-                    # width/height will be coded width/height x upscale factor
-                    original_width = video_info['streams'][video_stream_index]['width']
-                    original_height = video_info['streams'][video_stream_index]['height']
-                    scale_width = int(self.scale_ratio * original_width)
-                    scale_height = int(self.scale_ratio * original_height)
-
-                    # upscale images one by one using waifu2x
-                    Avalon.info(_('Starting to upscale extracted images'))
-                    self._upscale_frames()
-                    Avalon.info(_('Upscaling completed'))
-
-                    # frames to Video
-                    Avalon.info(_('Converting extracted frames into video'))
-
-                    # use user defined output size
-                    self.process_pool.append(fm.assemble_video(framerate,
-                                                               f'{scale_width}x{scale_height}',
-                                                               self.upscaled_frames))
-                    self._wait()
-                    Avalon.info(_('Conversion completed'))
-
-                    try:
-                        # migrate audio tracks and subtitles
-                        Avalon.info(_('Migrating audio, subtitles and other streams to upscaled video'))
-                        self.process_pool.append(fm.migrate_streams(self.current_input_video,
-                                                                    output_video,
-                                                                    self.upscaled_frames))
-                        self._wait()
-
-                    # if failed to copy streams
-                    # use file with only video stream
-                    except subprocess.CalledProcessError:
-                        Avalon.error(_('Failed to migrate streams'))
-                        Avalon.warning(_('Trying to output video without additional streams'))
-
-                        # construct output file path
-                        output_video_path = output_video.parent / f'{output_video.stem}{fm.intermediate_file_name.suffix}'
-
-                        # if output file already exists, cancel
-                        if output_video_path.exists():
-                            Avalon.error(_('Output video file exists, aborting'))
-
-                        # otherwise, rename intermediate file to the output file
-                        else:
-                            Avalon.info(_('Writing intermediate file to: {}').format(output_video_path.absolute()))
-                            (self.upscaled_frames / fm.intermediate_file_name).rename(output_video_path)
-
-                    # destroy temp directories
-                    self.cleanup_temp_directories()
-
-                except (Exception, KeyboardInterrupt, SystemExit) as e:
-                    with contextlib.suppress(ValueError):
-                        self.cleanup_temp_directories()
-                        self.running = False
-                    raise e
-
-            # increment total number of videos processed
-            self.total_processed += 1
-
-        # signal upscaling completion
-        self.running = False

src/video2x.c

@@ -0,0 +1,325 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include <libavutil/pixdesc.h>
+#include <libavutil/pixfmt.h>
+
+#include <libvideo2x.h>
+
+#include "getopt.h"
+
+const char *VIDEO2X_VERSION = "6.0.0";
+
+// Define command line options
+static struct option long_options[] = {
+    // General options
+    {"input", required_argument, NULL, 'i'},
+    {"output", required_argument, NULL, 'o'},
+    {"filter", required_argument, NULL, 'f'},
+    {"version", no_argument, NULL, 'v'},
+    {"help", no_argument, NULL, 0},
+
+    // Encoder options
+    {"codec", required_argument, NULL, 'c'},
+    {"preset", required_argument, NULL, 'p'},
+    {"pixfmt", required_argument, NULL, 'x'},
+    {"bitrate", required_argument, NULL, 'b'},
+    {"crf", required_argument, NULL, 'q'},
+
+    // Libplacebo options
+    {"shader", required_argument, NULL, 's'},
+    {"width", required_argument, NULL, 'w'},
+    {"height", required_argument, NULL, 'h'},
+
+    // RealESRGAN options
+    {"gpuid", required_argument, NULL, 'g'},
+    {"model", required_argument, NULL, 'm'},
+    {"scale", required_argument, NULL, 'r'},
+    {0, 0, 0, 0}
+};
+
+// Structure to hold parsed arguments
+struct arguments {
+    // General options
+    const char *input_filename;
+    const char *output_filename;
+    const char *filter_type;
+
+    // Encoder options
+    const char *codec;
+    const char *pix_fmt;
+    const char *preset;
+    int64_t bitrate;
+    float crf;
+
+    // libplacebo options
+    const char *shader_path;
+    int output_width;
+    int output_height;
+
+    // RealESRGAN options
+    int gpuid;
+    const char *model;
+    int scaling_factor;
+};
+
+const char *valid_models[] = {
+    "realesrgan-plus",
+    "realesrgan-plus-anime",
+    "realesr-animevideov3",
+};
+
+int is_valid_realesrgan_model(const char *model) {
+    if (!model) {
+        return 0;
+    }
+    for (int i = 0; i < sizeof(valid_models) / sizeof(valid_models[0]); i++) {
+        if (strcmp(model, valid_models[i]) == 0) {
+            return 1;
+        }
+    }
+    return 0;
+}
+
+void print_help() {
+    printf("Usage: video2x [OPTIONS]\n");
+    printf("\nGeneral Options:\n");
+    printf("  -i, --input		Input video file path\n");
+    printf("  -o, --output		Output video file path\n");
+    printf("  -f, --filter		Filter to use: 'libplacebo' or 'realesrgan'\n");
+    printf("  -v, --version		Print program version\n");
+    printf("  --help		Display this help page\n");
+
+    printf("\nEncoder Options (Optional):\n");
+    printf("  -c, --codec		Output codec (default: libx264)\n");
+    printf("  -p, --preset		Encoder preset (default: veryslow)\n");
+    printf("  -x, --pixfmt		Output pixel format (default: yuv420p)\n");
+    printf("  -b, --bitrate		Bitrate in bits per second (default: 2000000)\n");
+    printf("  -q, --crf		Constant Rate Factor (default: 17.0)\n");
+
+    printf("\nlibplacebo Options:\n");
+    printf("  -s, --shader		Name or path to custom GLSL shader file\n");
+    printf("  -w, --width		Output width\n");
+    printf("  -h, --height		Output height\n");
+
+    printf("\nRealESRGAN Options:\n");
+    printf("  -g, --gpuid		Vulkan GPU ID (default: 0)\n");
+    printf("  -m, --model		Name of the model to use\n");
+    printf("  -r, --scale		Scaling factor (2, 3, or 4)\n");
+}
+
+void parse_arguments(int argc, char **argv, struct arguments *arguments) {
+    int option_index = 0;
+    int c;
+
+    // Default argument values
+    arguments->input_filename = NULL;
+    arguments->output_filename = NULL;
+    arguments->filter_type = NULL;
+
+    // Encoder options
+    arguments->codec = "libx264";
+    arguments->preset = "veryslow";
+    arguments->pix_fmt = "yuv420p";
+    arguments->bitrate = 2 * 1000 * 1000;
+    arguments->crf = 17.0;
+
+    // libplacebo options
+    arguments->shader_path = NULL;
+    arguments->output_width = 0;
+    arguments->output_height = 0;
+
+    // RealESRGAN options
+    arguments->gpuid = 0;
+    arguments->model = NULL;
+    arguments->scaling_factor = 0;
+
+    while ((c = getopt_long(argc, argv, "i:o:f:c:x:p:b:q:s:w:h:r:m:v", long_options, &option_index)
+           ) != -1) {
+        switch (c) {
+            case 'i':
+                arguments->input_filename = optarg;
+                break;
+            case 'o':
+                arguments->output_filename = optarg;
+                break;
+            case 'f':
+                arguments->filter_type = optarg;
+                break;
+            case 'c':
+                arguments->codec = optarg;
+                break;
+            case 'x':
+                arguments->pix_fmt = optarg;
+                break;
+            case 'p':
+                arguments->preset = optarg;
+                break;
+            case 'b':
+                arguments->bitrate = strtoll(optarg, NULL, 10);
+                if (arguments->bitrate <= 0) {
+                    fprintf(stderr, "Error: Invalid bitrate specified.\n");
+                    exit(1);
+                }
+                break;
+            case 'q':
+                arguments->crf = atof(optarg);
+                if (arguments->crf < 0.0 || arguments->crf > 51.0) {
+                    fprintf(stderr, "Error: CRF must be between 0 and 51.\n");
+                    exit(1);
+                }
+                break;
+            case 's':
+                arguments->shader_path = optarg;
+                break;
+            case 'w':
+                arguments->output_width = atoi(optarg);
+                if (arguments->output_width <= 0) {
+                    fprintf(stderr, "Error: Output width must be greater than 0.\n");
+                    exit(1);
+                }
+                break;
+            case 'h':
+                arguments->output_height = atoi(optarg);
+                if (arguments->output_height <= 0) {
+                    fprintf(stderr, "Error: Output height must be greater than 0.\n");
+                    exit(1);
+                }
+                break;
+            case 'g':
+                arguments->gpuid = atoi(optarg);
+                break;
+            case 'm':
+                arguments->model = optarg;
+                if (!is_valid_realesrgan_model(arguments->model)) {
+                    fprintf(
+                        stderr,
+                        "Error: Invalid model specified. Must be 'realesrgan-plus', 'realesrgan-plus-anime', or 'realesr-animevideov3'.\n"
+                    );
+                    exit(1);
+                }
+                break;
+            case 'r':
+                arguments->scaling_factor = atoi(optarg);
+                if (arguments->scaling_factor != 2 && arguments->scaling_factor != 3 &&
+                    arguments->scaling_factor != 4) {
+                    fprintf(stderr, "Error: Scaling factor must be 2, 3, or 4.\n");
+                    exit(1);
+                }
+                break;
+            case 'v':
+                printf("video2x %s\n", VIDEO2X_VERSION);
+                exit(0);
+            case 0:  // Long-only options without short equivalents (e.g., help)
+                if (strcmp(long_options[option_index].name, "help") == 0) {
+                    print_help();
+                    exit(0);
+                }
+                break;
+            default:
+                fprintf(stderr, "Invalid options provided.\n");
+                exit(1);
+        }
+    }
+
+    // Check for required arguments
+    if (!arguments->input_filename || !arguments->output_filename) {
+        fprintf(stderr, "Error: Input and output files are required.\n");
+        exit(1);
+    }
+
+    if (!arguments->filter_type) {
+        fprintf(stderr, "Error: Filter type is required (libplacebo or realesrgan).\n");
+        exit(1);
+    }
+
+    if (strcmp(arguments->filter_type, "libplacebo") == 0) {
+        if (!arguments->shader_path || arguments->output_width == 0 ||
+            arguments->output_height == 0) {
+            fprintf(
+                stderr,
+                "Error: For libplacebo, shader name/path (-s), width (-w), and height (-e) are required.\n"
+            );
+            exit(1);
+        }
+    } else if (strcmp(arguments->filter_type, "realesrgan") == 0) {
+        if (arguments->scaling_factor == 0 || !arguments->model) {
+            fprintf(
+                stderr, "Error: For realesrgan, scaling factor (-r) and model (-m) are required.\n"
+            );
+            exit(1);
+        }
+    }
+}
+
+int main(int argc, char **argv) {
+    struct arguments arguments;
+    parse_arguments(argc, argv, &arguments);
+
+    // Setup filter configurations based on the parsed arguments
+    struct FilterConfig filter_config;
+    if (strcmp(arguments.filter_type, "libplacebo") == 0) {
+        filter_config.filter_type = FILTER_LIBPLACEBO;
+        filter_config.config.libplacebo.output_width = arguments.output_width;
+        filter_config.config.libplacebo.output_height = arguments.output_height;
+        filter_config.config.libplacebo.shader_path = arguments.shader_path;
+    } else if (strcmp(arguments.filter_type, "realesrgan") == 0) {
+        filter_config.filter_type = FILTER_REALESRGAN;
+        filter_config.config.realesrgan.gpuid = arguments.gpuid;
+        filter_config.config.realesrgan.tta_mode = 0;
+        filter_config.config.realesrgan.scaling_factor = arguments.scaling_factor;
+        filter_config.config.realesrgan.model = arguments.model;
+    } else {
+        fprintf(stderr, "Error: Invalid filter type specified.\n");
+        return 1;
+    }
+
+    // Parse codec to AVCodec
+    const AVCodec *codec = avcodec_find_encoder_by_name(arguments.codec);
+    if (!codec) {
+        fprintf(stderr, "Error: Codec '%s' not found.\n", arguments.codec);
+        return 1;
+    }
+
+    // Parse pixel format to AVPixelFormat
+    enum AVPixelFormat pix_fmt = av_get_pix_fmt(arguments.pix_fmt);
+    if (pix_fmt == AV_PIX_FMT_NONE) {
+        fprintf(stderr, "Error: Invalid pixel format '%s'.\n", arguments.pix_fmt);
+        return 1;
+    }
+
+    // Setup encoder configuration
+    struct EncoderConfig encoder_config = {
+        .output_width = 0,   // To be filled by libvideo2x
+        .output_height = 0,  // To be filled by libvideo2x
+        .codec = codec->id,
+        .pix_fmt = pix_fmt,
+        .preset = arguments.preset,
+        .bit_rate = arguments.bitrate,
+        .crf = arguments.crf,
+    };
+
+    // Setup struct to store processing status
+    struct ProcessingStatus status = {0};
+
+    // Process the video
+    if (process_video(
+            arguments.input_filename,
+            arguments.output_filename,
+            &filter_config,
+            &encoder_config,
+            &status
+        )) {
+        fprintf(stderr, "Video processing failed.\n");
+        return 1;
+    }
+
+    // Print processing summary
+    printf("====== Video2X Processing summary ======\n");
+    printf("Video file processed: %s\n", arguments.input_filename);
+    printf("Total frames processed: %ld\n", status.processed_frames);
+    printf("Total time taken: %lds\n", time(NULL) - status.start_time);
+    printf("Output written to: %s\n", arguments.output_filename);
+    return 0;
+}

src/video2x.pot

@@ -1,262 +0,0 @@
-# SOME DESCRIPTIVE TITLE.
-# Copyright (C) YEAR ORGANIZATION
-# FIRST AUTHOR <EMAIL@ADDRESS>, YEAR.
-#
-msgid ""
-msgstr ""
-"Project-Id-Version: PACKAGE VERSION\n"
-"POT-Creation-Date: 2020-05-11 04:39-0400\n"
-"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
-"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
-"Language-Team: LANGUAGE <LL@li.org>\n"
-"MIME-Version: 1.0\n"
-"Content-Type: text/plain; charset=cp1252\n"
-"Content-Transfer-Encoding: 8bit\n"
-"Generated-By: pygettext.py 1.5\n"
-
-
-#: progress_monitor.py:42
-msgid "Upscaling Progress"
-msgstr ""
-
-#: upscaler.py:104
-msgid "Specified or default cache directory is a file/link"
-msgstr ""
-
-#: upscaler.py:110
-msgid "Creating cache directory {}"
-msgstr ""
-
-#: upscaler.py:113
-msgid "Unable to create {}"
-msgstr ""
-
-#: upscaler.py:118
-msgid "Extracted frames are being saved to: {}"
-msgstr ""
-
-#: upscaler.py:120
-msgid "Upscaled frames are being saved to: {}"
-msgstr ""
-
-#: upscaler.py:130
-msgid "Cleaning up cache directory: {}"
-msgstr ""
-
-#: upscaler.py:133
-msgid "Unable to delete: {}"
-msgstr ""
-
-#: upscaler.py:139 upscaler.py:154 upscaler.py:165
-msgid "Input and output path type mismatch"
-msgstr ""
-
-#: upscaler.py:140
-msgid "Input is multiple files but output is not directory"
-msgstr ""
-
-#: upscaler.py:144
-msgid "Input path {} is neither a file nor a directory"
-msgstr ""
-
-#: upscaler.py:148 upscaler.py:170
-msgid "Input directory and output directory cannot be the same"
-msgstr ""
-
-#: upscaler.py:155
-msgid "Input is single file but output is directory"
-msgstr ""
-
-#: upscaler.py:158
-msgid "No suffix found in output file path"
-msgstr ""
-
-#: upscaler.py:159
-msgid "Suffix must be specified for FFmpeg"
-msgstr ""
-
-#: upscaler.py:166
-msgid "Input is directory but output is existing single file"
-msgstr ""
-
-#: upscaler.py:175
-msgid "Input path is neither a file nor a directory"
-msgstr ""
-
-#: upscaler.py:184
-msgid "FFmpeg or FFprobe cannot be found under the specified path"
-msgstr ""
-
-#: upscaler.py:185 upscaler.py:195
-msgid "Please check the configuration file settings"
-msgstr ""
-
-#: upscaler.py:194
-msgid "Specified driver executable directory doesn't exist"
-msgstr ""
-
-#: upscaler.py:221
-msgid "Failed to parse driver argument: {}"
-msgstr ""
-
-#: upscaler.py:253
-msgid "Unrecognized driver: {}"
-msgstr ""
-
-#: upscaler.py:293
-msgid "Starting progress monitor"
-msgstr ""
-
-#: upscaler.py:298
-msgid "Starting upscaled image cleaner"
-msgstr ""
-
-#: upscaler.py:307 upscaler.py:324
-msgid "Killing progress monitor"
-msgstr ""
-
-#: upscaler.py:310 upscaler.py:327
-msgid "Killing upscaled image cleaner"
-msgstr ""
-
-#: upscaler.py:331
-msgid "Terminating all processes"
-msgstr ""
-
-#: upscaler.py:338
-msgid "Main process waiting for subprocesses to exit"
-msgstr ""
-
-#: upscaler.py:357 upscaler.py:361
-msgid "Subprocess {} exited with code {}"
-msgstr ""
-
-#: upscaler.py:367
-msgid "Stop signal received"
-msgstr ""
-
-#: upscaler.py:372
-msgid "Subprocess execution ran into an error"
-msgstr ""
-
-#: upscaler.py:421
-msgid "Upscaling single video file: {}"
-msgstr ""
-
-#: upscaler.py:443 upscaler.py:502
-msgid "Starting to upscale extracted images"
-msgstr ""
-
-#: upscaler.py:448 upscaler.py:504
-msgid "Upscaling completed"
-msgstr ""
-
-#: upscaler.py:457
-msgid "Reading video information"
-msgstr ""
-
-#: upscaler.py:471
-msgid "Aborting: No video stream found"
-msgstr ""
-
-#: upscaler.py:490
-msgid "Unsupported pixel format: {}"
-msgstr ""
-
-#: upscaler.py:493
-msgid "Framerate: {}"
-msgstr ""
-
-#: upscaler.py:507
-msgid "Converting extracted frames into video"
-msgstr ""
-
-#: upscaler.py:514
-msgid "Conversion completed"
-msgstr ""
-
-#: upscaler.py:518
-msgid "Migrating audio, subtitles and other streams to upscaled video"
-msgstr ""
-
-#: upscaler.py:527
-msgid "Failed to migrate streams"
-msgstr ""
-
-#: upscaler.py:528
-msgid "Trying to output video without additional streams"
-msgstr ""
-
-#: upscaler.py:535
-msgid "Output video file exists, aborting"
-msgstr ""
-
-#: upscaler.py:539
-msgid "Writing intermediate file to: {}"
-msgstr ""
-
-#: video2x.py:84
-msgid ""
-"Video2X Version: {}\n"
-"Author: K4YT3X\n"
-"License: GNU GPL v3\n"
-"Github Page: https://github.com/k4yt3x/video2x\n"
-"Contact: k4yt3x@k4yt3x.com"
-msgstr ""
-
-#: video2x.py:106
-msgid "Video2X Options"
-msgstr ""
-
-#: video2x.py:107
-msgid "show this help message and exit"
-msgstr ""
-
-#: video2x.py:108
-msgid "source video file/directory"
-msgstr ""
-
-#: video2x.py:109
-msgid "output video file/directory"
-msgstr ""
-
-#: video2x.py:110
-msgid "video2x config file path"
-msgstr ""
-
-#: video2x.py:112
-msgid "display version, lawful information and exit"
-msgstr ""
-
-#: video2x.py:115
-msgid "Upscaling Options"
-msgstr ""
-
-#: video2x.py:116
-msgid "upscaling driver"
-msgstr ""
-
-#: video2x.py:117
-msgid "scaling ratio"
-msgstr ""
-
-#: video2x.py:118
-msgid "number of processes to use for upscaling"
-msgstr ""
-
-#: video2x.py:119
-msgid "preserve extracted and upscaled frames"
-msgstr ""
-
-#: video2x.py:159
-msgid "This file cannot be imported"
-msgstr ""
-
-#: video2x.py:229
-msgid "Program completed, taking {} seconds"
-msgstr ""
-
-#: video2x.py:232
-msgid "An exception has occurred"
-msgstr ""
-

src/video2x.py

@@ -1,233 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-r"""
-
-__      __  _       _                  ___   __   __
-\ \    / / (_)     | |                |__ \  \ \ / /
- \ \  / /   _    __| |   ___    ___      ) |  \ V /
-  \ \/ /   | |  / _` |  / _ \  / _ \    / /    > <
-   \  /    | | | (_| | |  __/ | (_) |  / /_   / . \
-    \/     |_|  \__,_|  \___|  \___/  |____| /_/ \_\
-
-
-Name: Video2X Controller
-Creator: K4YT3X
-Date Created: Feb 24, 2018
-Last Modified: May 10, 2020
-
-Editor: BrianPetkovsek
-Last Modified: June 17, 2019
-
-Editor: SAT3LL
-Last Modified: June 25, 2019
-
-Editor: 28598519a
-Last Modified: March 23, 2020
-
-Licensed under the GNU General Public License Version 3 (GNU GPL v3),
-    available at: https://www.gnu.org/licenses/gpl-3.0.txt
-
-(C) 2018 - 2020 K4YT3X
-
-Video2X is free software: you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation, either version 3 of the License, or
-(at your option) any later version.
-
-Video2X is distributed in the hope that it will be useful,
-but WITHOUT ANY WARRANTY; without even the implied warranty of
-MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-GNU General Public License for more details.
-
-You should have received a copy of the GNU General Public License
-along with this program. If not, see <https://www.gnu.org/licenses/>.
-
-Description: Video2X is an automation software based on waifu2x image
-enlarging engine. It extracts frames from a video, enlarge it by a
-number of times without losing any details or quality, keeping lines
-smooth and edges sharp.
-"""
-
-# local imports
-from upscaler import AVAILABLE_DRIVERS
-from upscaler import Upscaler
-
-# built-in imports
-import argparse
-import gettext
-import importlib
-import locale
-import os
-import pathlib
-import sys
-import tempfile
-import time
-import traceback
-import yaml
-
-# third-party imports
-from avalon_framework import Avalon
-
-# internationalization constants
-DOMAIN = 'video2x'
-LOCALE_DIRECTORY = pathlib.Path(__file__).parent.absolute() / 'locale'
-
-# getting default locale settings
-default_locale, encoding = locale.getdefaultlocale()
-language = gettext.translation(DOMAIN, LOCALE_DIRECTORY, [default_locale], fallback=True)
-language.install()
-_ = language.gettext
-
-
-VERSION = '4.0.0'
-
-LEGAL_INFO = _('''Video2X Version: {}
-Author: K4YT3X
-License: GNU GPL v3
-Github Page: https://github.com/k4yt3x/video2x
-Contact: k4yt3x@k4yt3x.com''').format(VERSION)
-
-LOGO = r'''
-    __      __  _       _                  ___   __   __
-    \ \    / / (_)     | |                |__ \  \ \ / /
-     \ \  / /   _    __| |   ___    ___      ) |  \ V /
-      \ \/ /   | |  / _` |  / _ \  / _ \    / /    > <
-       \  /    | | | (_| | |  __/ | (_) |  / /_   / . \
-        \/     |_|  \__,_|  \___|  \___/  |____| /_/ \_\
-'''
-
-
-def parse_arguments():
-    """ parse CLI arguments
-    """
-    parser = argparse.ArgumentParser(prog='video2x', formatter_class=argparse.ArgumentDefaultsHelpFormatter, add_help=False)
-
-    # video options
-    video2x_options = parser.add_argument_group(_('Video2X Options'))
-    video2x_options.add_argument('-h', '--help', action='help', help=_('show this help message and exit'))
-    video2x_options.add_argument('-i', '--input', type=pathlib.Path, help=_('source video file/directory'))
-    video2x_options.add_argument('-o', '--output', type=pathlib.Path, help=_('output video file/directory'))
-    video2x_options.add_argument('-c', '--config', type=pathlib.Path, help=_('video2x config file path'), action='store',
-                                 default=pathlib.Path(__file__).parent.absolute() / 'video2x.yaml')
-    video2x_options.add_argument('-v', '--version', help=_('display version, lawful information and exit'), action='store_true')
-
-    # scaling options
-    upscaling_options = parser.add_argument_group(_('Upscaling Options'))
-    upscaling_options.add_argument('-d', '--driver', help=_('upscaling driver'), choices=AVAILABLE_DRIVERS, default='waifu2x_caffe')
-    upscaling_options.add_argument('-r', '--ratio', help=_('scaling ratio'), action='store', type=float, default=2.0)
-    upscaling_options.add_argument('-p', '--processes', help=_('number of processes to use for upscaling'), action='store', type=int, default=1)
-    upscaling_options.add_argument('--preserve_frames', help=_('preserve extracted and upscaled frames'), action='store_true')
-
-    # if no driver arguments are specified
-    if '--' not in sys.argv:
-        video2x_args = parser.parse_args()
-        return video2x_args, None
-
-    # if driver arguments are specified
-    else:
-        video2x_args = parser.parse_args(sys.argv[1:sys.argv.index('--')])
-        wrapper = getattr(importlib.import_module(f'wrappers.{video2x_args.driver}'), 'WrapperMain')
-        driver_args = wrapper.parse_arguments(sys.argv[sys.argv.index('--') + 1:])
-        return video2x_args, driver_args
-
-
-def print_logo():
-    """print video2x logo"""
-    print(LOGO)
-    print(f'\n{"Video2X Video Enlarger".rjust(40, " ")}')
-    print(f'\n{Avalon.FM.BD}{f"Version {VERSION}".rjust(36, " ")}{Avalon.FM.RST}\n')
-
-
-def read_config(config_file: pathlib.Path) -> dict:
-    """ read video2x configurations from config file
-
-    Arguments:
-        config_file {pathlib.Path} -- video2x configuration file pathlib.Path
-
-    Returns:
-        dict -- dictionary of video2x configuration
-    """
-
-    with open(config_file, 'r') as config:
-        return yaml.load(config, Loader=yaml.FullLoader)
-
-
-# /////////////////// Execution /////////////////// #
-
-# this is not a library
-if __name__ != '__main__':
-    Avalon.error(_('This file cannot be imported'))
-    raise ImportError(f'{__file__} cannot be imported')
-
-# print video2x logo
-print_logo()
-
-# parse command line arguments
-video2x_args, driver_args = parse_arguments()
-
-# display version and lawful informaition
-if video2x_args.version:
-    print(LEGAL_INFO)
-    sys.exit(0)
-
-# read configurations from configuration file
-config = read_config(video2x_args.config)
-
-# load waifu2x configuration
-driver_settings = config[video2x_args.driver]
-driver_settings['path'] = os.path.expandvars(driver_settings['path'])
-
-# read FFmpeg configuration
-ffmpeg_settings = config['ffmpeg']
-ffmpeg_settings['ffmpeg_path'] = os.path.expandvars(ffmpeg_settings['ffmpeg_path'])
-
-# load video2x settings
-image_format = config['video2x']['image_format'].lower()
-preserve_frames = config['video2x']['preserve_frames']
-
-# if preserve frames specified in command line
-# overwrite config file options
-if video2x_args.preserve_frames is True:
-    preserve_frames = True
-
-# if cache directory not specified
-# use default path: %TEMP%\video2x
-if config['video2x']['video2x_cache_directory'] is None:
-    video2x_cache_directory = (pathlib.Path(tempfile.gettempdir()) / 'video2x')
-else:
-    video2x_cache_directory = pathlib.Path(config['video2x']['video2x_cache_directory'])
-
-# overwrite driver_settings with driver_args
-if driver_args is not None:
-    driver_args_dict = vars(driver_args)
-    for key in driver_args_dict:
-        if driver_args_dict[key] is not None:
-            driver_settings[key] = driver_args_dict[key]
-
-# start execution
-try:
-    # start timer
-    begin_time = time.time()
-
-    # initialize upscaler object
-    upscaler = Upscaler(input_path=video2x_args.input,
-                        output_path=video2x_args.output,
-                        driver_settings=driver_settings,
-                        ffmpeg_settings=ffmpeg_settings)
-
-    # set upscaler optional options
-    upscaler.driver = video2x_args.driver
-    upscaler.scale_ratio = video2x_args.ratio
-    upscaler.processes = video2x_args.processes
-    upscaler.video2x_cache_directory = video2x_cache_directory
-    upscaler.image_format = image_format
-    upscaler.preserve_frames = preserve_frames
-
-    # run upscaler
-    upscaler.run()
-
-    Avalon.info(_('Program completed, taking {} seconds').format(round((time.time() - begin_time), 5)))
-
-except Exception:
-    Avalon.error(_('An exception has occurred'))
-    traceback.print_exc()

src/video2x.yaml

@@ -1,144 +0,0 @@
-# Name: Video2X Configuration File
-# Creator: K4YT3X
-# Date Created: October 23, 2018
-# Last Modified: May 9, 2020
-# Values here are the default values. Change the value here to
-#   save the default value permanently.
-# Items commented out are parameters irrelevant to this context
-#   or parameters handled by Video2X internally.
-waifu2x_caffe:
-  path: '%LOCALAPPDATA%\video2x\waifu2x-caffe\waifu2x-caffe-cui'
-  tta: 0 # <0|1> 8x slower and slightly high quality
-  gpu: 0 # gpu device no
-  batch_size: 1 # input batch size
-  crop_h: null # input image split size(height)
-  crop_w: null # input image split size(width)
-  crop_size: 128 # input image split size
-  output_depth: 8 # output image chanel depth bit
-  output_quality: -1 # output image quality
-  process: gpu # <cpu|gpu|cudnn> process mode
-  model_dir: null # path to custom model directory (don't append last / )
-  scale_height: 0 # custom scale height (specifying this will overwrite scale_ratio)
-  scale_width: 0 # custom scale width (specifying this will overwrite scale_ratio)
-  #scale_ratio: null # custom scale ratio
-  noise_level: 3 # <0|1|2|3> noise reduction level
-  mode: noise_scale # <noise|scale|noise_scale|auto_scale> image processing mode
-  output_extention: null # extension to output image file when output_path is (auto) or input_path is folder
-  input_extention_list: null # extension to input image file when input_path is folder
-  #output_path: null # path to output image file (when input_path is folder, output_path must be folder)
-  #input_path: null # (required) path to input image file
-waifu2x_converter_cpp:
-  path: '%LOCALAPPDATA%\video2x\waifu2x-converter-cpp\waifu2x-converter-cpp'
-  #list-supported-formats: null # dump currently supported format list
-  #list-opencv-formats: null # (deprecated. Use --list-supported-formats) dump opencv supported format list
-  #list-processor # dump processor list
-  output-format: null # The format used when running in recursive/folder mode
-  png-compression: 5 # Set PNG compression level (0-9), 9 = Max compression (slowest & smallest)
-  image-quality: -1 # JPEG & WebP Compression quality (-1-101, 0 being smallest size and lowest quality, -1 being default), use 101 for lossless WebP
-  block-size: 0 # block size
-  disable-gpu: false # disable GPU
-  force-OpenCL: false # force to use OpenCL on Intel Platform
-  processor: -1 # set target processor (-1 uses default device)
-  jobs: 0 # number of threads launching at the same time
-  model-dir: null # path to custom model directory (don't append last / ) default: models_rgb
-  #scale-ratio: 2.0 # custom scale ratio
-  noise-level: 1 # <0|1|2|3> noise reduction level
-  mode: noise-scale # <noise|scale|noise-scale> image processing mode
-  log-level: 1 # <0|1|2|3|4> Set log level
-  silent: true # Enable silent mode. (same as --log-level 1)
-  tta: 0 # Enable Test-Time Augmentation mode. (0 or 1)
-  #generate-subdir: 0 # Generate sub folder when recursive directory is enabled.
-  #auto-naming: 0 # Add postfix to output name when output path is not specified.
-  #recursive-directory: 0 # Search recursively through directories to find more images to process.
-  #output: null # path to output image file or directory  (you should use the full path)
-  #input: null # (required) path to input image file or directory (you should use the full path)
-waifu2x_ncnn_vulkan:
-  path: '%LOCALAPPDATA%\video2x\waifu2x-ncnn-vulkan\waifu2x-ncnn-vulkan'
-  v: null # verbose output
-  #i: null # input-path: input image path (jpg/png) or directory
-  #o: null # output-path: output image path (png) or directory
-  'n': 2 # noise-level: denoise level (-1/0/1/2/3, default=0)
-  s: 2 # scale: upscale ratio (1/2, default=2)
-  t: 400 # tile-size: tile size (>=32, default=400)
-  m: null # model-path: waifu2x model path (default=models-cunet)
-  g: 0 # gpu-id: gpu device to use (default=0)
-  j: '1:2:2' # thread count for load/proc/save (default=1:2:2)
-  x: false # enable tta mode
-srmd_ncnn_vulkan:
-  path: '%LOCALAPPDATA%\video2x\srmd-ncnn-vulkan\srmd-ncnn-vulkan'
-  v: null # verbose output
-  #i: null # input-path: input image path (jpg/png) or directory
-  #o: null # output-path: output image path (png) or directory
-  'n': 3 # noise-level: denoise level (-1/0/1/2/3/4/5/6/7/8/9/10, default=3)
-  s: 2 # upscale ratio (2/3/4, default=2)
-  t: 400 # tile-size: tile size (>=32, default=400)
-  m: null # srmd model path (default=models-srmd)
-  g: 0 # gpu device to use (default=0)
-  j: '1:2:2' # thread count for load/proc/save (default=1:2:2)
-  x: false # enable tta mode
-anime4kcpp:
-  path: '%LOCALAPPDATA%\video2x\anime4kcpp\CLI\Anime4KCPP_CLI\Anime4KCPP_CLI'
-  #input: null # File for loading (string [=./pic/p1.png])
-  #output: null # File for outputting (string [=output.png])
-  passes: 2 # Passes for processing (int [=2])
-  pushColorCount: 2 # Limit the number of color pushes (int [=2])
-  strengthColor: 0.3 # Strength for pushing color,range 0 to 1,higher for thinner (double [=0.3])
-  strengthGradient: 1.0 # Strength for pushing gradient,range 0 to 1,higher for sharper (double [=1])
-  zoomFactor: 2.0 # zoom factor for resizing (double [=2])
-  threads: 16 # Threads count for video processing (unsigned int [=16])
-  fastMode: false # Faster but maybe low quality
-  videoMode: true # Video process
-  preview: null # Preview image
-  preprocessing: False # Enable pre processing
-  postprocessing: False # Enable post processing
-  preFilters: 4 # Enhancement filter, only working when preProcessing is true,there are 5 options by binary:Median blur=0000001, Mean blur=0000010, CAS Sharpening=0000100, Gaussian blur weak=0001000, Gaussian blur=0010000, Bilateral filter=0100000, Bilateral filter faster=1000000, you can freely combine them, eg: Gaussian blur weak + Bilateral filter = 0001000 | 0100000 = 0101000 = 40(D) (unsigned int [=4])
-  postFilters: 40 # Enhancement filter, only working when postProcessing is true,there are 5 options by binary:Median blur=0000001, Mean blur=0000010, CAS Sharpening=0000100, Gaussian blur weak=0001000, Gaussian blur=0010000, Bilateral filter=0100000, Bilateral filter faster=1000000, you can freely combine them, eg: Gaussian blur weak + Bilateral filter = 0001000 | 0100000 = 0101000 = 40(D), so you can put 40 to enable Gaussian blur weak and Bilateral filter, which also is what I recommend for image that < 1080P, 48 for image that >= 1080P, and for performance I recommend to use 72 for video that < 1080P, 80 for video that >=1080P (unsigned int [=40])
-  GPUMode: False # Enable GPU acceleration
-  listGPUs: null # list GPUs
-  platformID: 0 # Specify the platform ID (unsigned int [=0])
-  deviceID: 0 # Specify the device ID (unsigned int [=0])
-  codec: mp4v # Specify the codec for encoding from mp4v(recommended in Windows), dxva(for Windows), avc1(H264, recommended in Linux), vp09(very slow), hevc(not support in Windowds), av01(not support in Windowds) (string [=mp4v])
-ffmpeg:
-  ffmpeg_path: '%LOCALAPPDATA%\video2x\ffmpeg-latest-win64-static\bin'
-  intermediate_file_name: 'intermediate.mkv'
-  # step 1: extract all frames from original video
-  # into temporary directory
-  video_to_frames:
-    output_options:
-      '-qscale:v': null
-      '-pix_fmt': rgba64be
-    '-hwaccel': auto
-    '-y': true
-  # step 2: stitch all frames back into a video
-  # with only a video track
-  frames_to_video:
-    input_options:
-      '-qscale:v': null
-      '-qscale:a': null
-      '-f': image2
-    output_options:
-      '-vcodec': libx264
-      '-crf': 17
-      '-b:v': null
-      '-pix_fmt': null
-    '-hwaccel': auto
-    '-y': true
-  # step 3: migrate audio and subtitle tracks from original
-  # video into the upscaled video
-  migrating_tracks:
-    output_options:
-      '-map':
-        - '0:v?'
-        - '1:a?'
-        - '1:s?'
-        - '1:d?'
-        - '1:t?'
-      '-c': copy
-      '-pix_fmt': null
-      '-metadata': 'comment=Upscaled by Video2X'
-    '-hwaccel': auto
-    '-y': true
-video2x:
-  video2x_cache_directory: null # default: %TEMP%\video2x
-  image_format: png
-  preserve_frames: false

src/video2x_gui.py

@@ -1,874 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Creator: Video2X GUI
-Author: K4YT3X
-Date Created: May 5, 2020
-Last Modified: May 10, 2020
-"""
-
-# local imports
-from upscaler import Upscaler
-
-# built-in imports
-import contextlib
-import os
-import pathlib
-import sys
-import tempfile
-import time
-import traceback
-import urllib
-import yaml
-
-# third-party imports
-from PyQt5 import QtGui, uic
-from PyQt5.QtCore import *
-from PyQt5.QtWidgets import *
-# QObject, pyqtSlot, pyqtSignal, QRunnable, QThreadPool, QAbstractTableModel, Qt
-
-VERSION = '2.0.0'
-
-LEGAL_INFO = f'''Video2X GUI Version: {VERSION}\\
-Author: K4YT3X\\
-License: GNU GPL v3\\
-Github Page: [https://github.com/k4yt3x/video2x](https://github.com/k4yt3x/video2x)\\
-Contact: [k4yt3x@k4yt3x.com](mailto:k4yt3x@k4yt3x.com)'''
-
-AVAILABLE_DRIVERS = {
-    'Waifu2X Caffe': 'waifu2x_caffe',
-    'Waifu2X Converter CPP': 'waifu2x_converter_cpp',
-    'Waifu2X NCNN Vulkan': 'waifu2x_ncnn_vulkan',
-    'SRMD NCNN Vulkan': 'srmd_ncnn_vulkan',
-    'Anime4KCPP': 'anime4kcpp'
-}
-
-
-def resource_path(relative_path: str) -> pathlib.Path:
-    try:
-        base_path = pathlib.Path(sys._MEIPASS)
-    except AttributeError:
-        base_path = pathlib.Path(__file__).parent
-    return base_path / relative_path
-
-
-class WorkerSignals(QObject):
-    progress = pyqtSignal(tuple)
-    error = pyqtSignal(Exception)
-    interrupted = pyqtSignal()
-    finished = pyqtSignal()
-
-
-class ProgressMonitorWorkder(QRunnable):
-    def __init__(self, fn, *args, **kwargs):
-        super(ProgressMonitorWorkder, self).__init__()
-        self.fn = fn
-        self.args = args
-        self.kwargs = kwargs
-        self.signals = WorkerSignals()
-        self.kwargs['progress_callback'] = self.signals.progress
-
-    @pyqtSlot()
-    def run(self):
-        try:
-            self.fn(*self.args, **self.kwargs)
-        except Exception:
-            pass
-
-
-class UpscalerWorker(QRunnable):
-
-    def __init__(self, fn, *args, **kwargs):
-        super(UpscalerWorker, self).__init__()
-
-        # Store constructor arguments (re-used for processing)
-        self.fn = fn
-        self.args = args
-        self.kwargs = kwargs
-        self.signals = WorkerSignals()
-
-    @pyqtSlot()
-    def run(self):
-
-        # Retrieve args/kwargs here; and fire processing using them
-        try:
-            self.fn(*self.args, **self.kwargs)
-        except (KeyboardInterrupt, SystemExit):
-            self.signals.interrupted.emit()
-        except Exception as e:
-            traceback.print_exc()
-            self.signals.error.emit(e)
-        else:
-            self.signals.finished.emit()
-
-
-class InputTableModel(QAbstractTableModel):
-    def __init__(self, data):
-        super(InputTableModel, self).__init__()
-        self._data = data
-
-    def data(self, index, role):
-        if role == Qt.DisplayRole:
-
-            if index.column() == 0:
-                return str(self._data[index.row()].absolute())
-            else:
-                if self._data[index.row()].is_file():
-                    return 'File'
-                elif self._data[index.row()].is_dir():
-                    return 'Folder'
-                else:
-                    return 'Unknown'
-
-    def rowCount(self, index):
-        return len(self._data)
-
-    def columnCount(self, index):
-        return 2
-
-    def removeRow(self, index):
-        self._data.pop(index)
-
-    def headerData(self, section, orientation, role):
-        # section is the index of the column/row.
-        if role != Qt.DisplayRole:
-            return None
-
-        horizontal_headers = ['File Path', 'Type']
-
-        # return the correspondign header
-        if orientation == Qt.Horizontal:
-            return horizontal_headers[section]
-
-        # simply return the line number
-        if orientation == Qt.Vertical:
-            return str(section)
-
-
-class Video2XMainWindow(QMainWindow):
-
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-        uic.loadUi(str(resource_path('video2x_gui.ui')), self)
-
-        # create thread pool for upscaler workers
-        self.threadpool = QThreadPool()
-
-        # set window title and icon
-        self.video2x_icon_path = str(resource_path('images/video2x.png'))
-        self.setWindowTitle(f'Video2X GUI {VERSION}')
-        self.setWindowIcon(QtGui.QIcon(self.video2x_icon_path))
-
-        # menu bar
-        self.action_exit = self.findChild(QAction, 'actionExit')
-        self.action_exit.triggered.connect(sys.exit)
-
-        self.action_about = self.findChild(QAction, 'actionAbout')
-        self.action_about.triggered.connect(self.show_about)
-
-        # main tab
-        # select input file/folder
-        self.input_table_view = self.findChild(QTableView, 'inputTableView')
-        self.input_table_view.dragEnterEvent = self.dragEnterEvent
-        self.input_table_view.dropEvent = self.dropEvent
-
-        # initialize data in table
-        self.input_table_data = []
-        self.input_table_model = InputTableModel(self.input_table_data)
-        self.input_table_view.setModel(self.input_table_model)
-        # stretch file path and fill columns horizontally
-        self.input_table_view.horizontalHeader().setSectionResizeMode(0, QHeaderView.Stretch)
-
-        # input table buttons
-        self.input_select_file_button = self.findChild(QPushButton, 'inputSelectFileButton')
-        self.input_select_file_button.clicked.connect(self.select_input_file)
-        self.input_select_folder_button = self.findChild(QPushButton, 'inputSelectFolderButton')
-        self.input_select_folder_button.clicked.connect(self.select_input_folder)
-        self.input_delete_selected_button = self.findChild(QPushButton, 'inputDeleteSelectedButton')
-        self.input_delete_selected_button.clicked.connect(self.input_table_delete_selected)
-        self.input_clear_all_button = self.findChild(QPushButton, 'inputClearAllButton')
-        self.input_clear_all_button.clicked.connect(self.input_table_clear_all)
-
-        # other paths selection
-        # select output file/folder
-        self.output_line_edit = self.findChild(QLineEdit, 'outputLineEdit')
-        self.enable_line_edit_file_drop(self.output_line_edit)
-        self.output_line_edit.setText(str((pathlib.Path().cwd() / 'output').absolute()))
-        self.output_select_file_button = self.findChild(QPushButton, 'outputSelectFileButton')
-        self.output_select_file_button.clicked.connect(self.select_output_file)
-        self.output_select_folder_button = self.findChild(QPushButton, 'outputSelectFolderButton')
-        self.output_select_folder_button.clicked.connect(self.select_output_folder)
-
-        # config file
-        self.config_line_edit = self.findChild(QLineEdit, 'configLineEdit')
-        self.enable_line_edit_file_drop(self.config_line_edit)
-        self.config_line_edit.setText(str((pathlib.Path(__file__).parent / 'video2x.yaml').absolute()))
-        self.config_select_file_button = self.findChild(QPushButton, 'configSelectButton')
-        self.config_select_file_button.clicked.connect(self.select_config_file)
-
-        # cache directory
-        self.cache_line_edit = self.findChild(QLineEdit, 'cacheLineEdit')
-        self.enable_line_edit_file_drop(self.cache_line_edit)
-        self.cache_select_folder_button = self.findChild(QPushButton, 'cacheSelectFolderButton')
-        self.cache_select_folder_button.clicked.connect(self.select_cache_folder)
-
-        # express settings
-        self.driver_combo_box = self.findChild(QComboBox, 'driverComboBox')
-        self.driver_combo_box.currentTextChanged.connect(self.update_gui_for_driver)
-        self.processes_spin_box = self.findChild(QSpinBox, 'processesSpinBox')
-        self.scale_ratio_double_spin_box = self.findChild(QDoubleSpinBox, 'scaleRatioDoubleSpinBox')
-        self.preserve_frames_check_box = self.findChild(QCheckBox, 'preserveFramesCheckBox')
-
-        # frame preview
-        self.frame_preview_show_preview_check_box = self.findChild(QCheckBox, 'framePreviewShowPreviewCheckBox')
-        self.frame_preview_keep_aspect_ratio_check_box = self.findChild(QCheckBox, 'framePreviewKeepAspectRatioCheckBox')
-        self.frame_preview_label = self.findChild(QLabel, 'framePreviewLabel')
-
-        # currently processing
-        self.currently_processing_label = self.findChild(QLabel, 'currentlyProcessingLabel')
-        self.current_progress_bar = self.findChild(QProgressBar, 'currentProgressBar')
-        self.time_elapsed_label = self.findChild(QLabel, 'timeElapsedLabel')
-        self.time_remaining_label = self.findChild(QLabel, 'timeRemainingLabel')
-        self.rate_label = self.findChild(QLabel, 'rateLabel')
-        self.frames_label = self.findChild(QLabel, 'framesLabel')
-
-        # overall progress
-        self.overall_progress_bar = self.findChild(QProgressBar, 'overallProgressBar')
-        self.overall_progress_label = self.findChild(QLabel, 'overallProgressLabel')
-        self.start_button = self.findChild(QPushButton, 'startButton')
-        self.start_button.clicked.connect(self.start)
-        self.stop_button = self.findChild(QPushButton, 'stopButton')
-        self.stop_button.clicked.connect(self.stop)
-
-        # driver settings
-        # waifu2x-caffe
-        self.waifu2x_caffe_path_line_edit = self.findChild(QLineEdit, 'waifu2xCaffePathLineEdit')
-        self.enable_line_edit_file_drop(self.waifu2x_caffe_path_line_edit)
-        self.waifu2x_caffe_path_select_button = self.findChild(QPushButton, 'waifu2xCaffePathSelectButton')
-        self.waifu2x_caffe_path_select_button.clicked.connect(lambda: self.select_driver_binary_path(self.waifu2x_caffe_path_line_edit))
-        self.waifu2x_caffe_scale_width_spin_box = self.findChild(QSpinBox, 'waifu2xCaffeScaleWidthSpinBox')
-        self.waifu2x_caffe_scale_height_spin_box = self.findChild(QSpinBox, 'waifu2xCaffeScaleHeightSpinBox')
-        self.waifu2x_caffe_mode_combo_box = self.findChild(QComboBox, 'waifu2xCaffeModeComboBox')
-        self.waifu2x_caffe_noise_level_spin_box = self.findChild(QSpinBox, 'waifu2xCaffeNoiseLevelSpinBox')
-        self.waifu2x_caffe_process_combo_box = self.findChild(QComboBox, 'waifu2xCaffeProcessComboBox')
-        self.waifu2x_caffe_model_combobox = self.findChild(QComboBox, 'waifu2xCaffeModelComboBox')
-        self.waifu2x_caffe_crop_size_spin_box = self.findChild(QSpinBox, 'waifu2xCaffeCropSizeSpinBox')
-        self.waifu2x_caffe_output_quality_spin_box = self.findChild(QSpinBox, 'waifu2xCaffeOutputQualitySpinBox')
-        self.waifu2x_caffe_output_depth_spin_box = self.findChild(QSpinBox, 'waifu2xCaffeOutputDepthSpinBox')
-        self.waifu2x_caffe_batch_size_spin_box = self.findChild(QSpinBox, 'waifu2xCaffeBatchSizeSpinBox')
-        self.waifu2x_caffe_gpu_spin_box = self.findChild(QSpinBox, 'waifu2xCaffeGpuSpinBox')
-        self.waifu2x_caffe_tta_check_box = self.findChild(QCheckBox, 'waifu2xCaffeTtaCheckBox')
-
-        # waifu2x-converter-cpp
-        self.waifu2x_converter_cpp_path_line_edit = self.findChild(QLineEdit, 'waifu2xConverterCppPathLineEdit')
-        self.enable_line_edit_file_drop(self.waifu2x_converter_cpp_path_line_edit)
-        self.waifu2x_converter_cpp_path_edit_button = self.findChild(QPushButton, 'waifu2xConverterCppPathSelectButton')
-        self.waifu2x_converter_cpp_path_edit_button.clicked.connect(lambda: self.select_driver_binary_path(self.waifu2x_converter_cpp_path_line_edit))
-        self.waifu2x_converter_cpp_png_compression_spin_box = self.findChild(QSpinBox, 'waifu2xConverterCppPngCompressionSpinBox')
-        self.waifu2x_converter_cpp_image_quality_spin_box = self.findChild(QSpinBox, 'waifu2xConverterCppImageQualitySpinBox')
-        self.waifu2x_converter_cpp_block_size_spin_box = self.findChild(QSpinBox, 'waifu2xConverterCppBlockSizeSpinBox')
-        self.waifu2x_converter_cpp_processor_spin_box = self.findChild(QSpinBox, 'waifu2xConverterCppProcessorSpinBox')
-        self.waifu2x_converter_cpp_model_combo_box = self.findChild(QComboBox, 'waifu2xConverterCppModelComboBox')
-        self.waifu2x_converter_cpp_noise_level_spin_box = self.findChild(QSpinBox, 'waifu2xConverterCppNoiseLevelSpinBox')
-        self.waifu2x_converter_cpp_mode_combo_box = self.findChild(QComboBox, 'waifu2xConverterCppModeComboBox')
-        self.waifu2x_converter_cpp_log_level_spin_box = self.findChild(QSpinBox, 'waifu2xConverterCppLogLevelSpinBox')
-        self.waifu2x_converter_cpp_disable_gpu_check_box = self.findChild(QCheckBox, 'waifu2xConverterCppDisableGpuCheckBox')
-        self.waifu2x_converter_cpp_force_opencl_check_box = self.findChild(QCheckBox, 'waifu2xConverterCppForceOpenclCheckBox')
-        self.waifu2x_converter_cpp_tta_check_box = self.findChild(QCheckBox, 'waifu2xConverterCppTtaCheckBox')
-
-        # waifu2x-ncnn-vulkan
-        self.waifu2x_ncnn_vulkan_path_line_edit = self.findChild(QLineEdit, 'waifu2xNcnnVulkanPathLineEdit')
-        self.enable_line_edit_file_drop(self.waifu2x_ncnn_vulkan_path_line_edit)
-        self.waifu2x_ncnn_vulkan_path_select_button = self.findChild(QPushButton, 'waifu2xNcnnVulkanPathSelectButton')
-        self.waifu2x_ncnn_vulkan_path_select_button.clicked.connect(lambda: self.select_driver_binary_path(self.waifu2x_ncnn_vulkan_path_line_edit))
-        self.waifu2x_ncnn_vulkan_noise_level_spin_box = self.findChild(QSpinBox, 'waifu2xNcnnVulkanNoiseLevelSpinBox')
-        self.waifu2x_ncnn_vulkan_tile_size_spin_box = self.findChild(QSpinBox, 'waifu2xNcnnVulkanTileSizeSpinBox')
-        self.waifu2x_ncnn_vulkan_model_combo_box = self.findChild(QComboBox, 'waifu2xNcnnVulkanModelComboBox')
-        self.waifu2x_ncnn_vulkan_gpu_id_spin_box = self.findChild(QSpinBox, 'waifu2xNcnnVulkanGpuIdSpinBox')
-        self.waifu2x_ncnn_vulkan_jobs_line_edit = self.findChild(QLineEdit, 'waifu2xNcnnVulkanJobsLineEdit')
-        self.waifu2x_ncnn_vulkan_tta_check_box = self.findChild(QCheckBox, 'waifu2xNcnnVulkanTtaCheckBox')
-
-        # srmd-ncnn-vulkan
-        self.srmd_ncnn_vulkan_path_line_edit = self.findChild(QLineEdit, 'srmdNcnnVulkanPathLineEdit')
-        self.enable_line_edit_file_drop(self.srmd_ncnn_vulkan_path_line_edit)
-        self.srmd_ncnn_vulkan_path_select_button = self.findChild(QPushButton, 'srmdNcnnVulkanPathSelectButton')
-        self.srmd_ncnn_vulkan_path_select_button.clicked.connect(lambda: self.select_driver_binary_path(self.srmd_ncnn_vulkan_path_line_edit))
-        self.srmd_ncnn_vulkan_noise_level_spin_box = self.findChild(QSpinBox, 'srmdNcnnVulkanNoiseLevelSpinBox')
-        self.srmd_ncnn_vulkan_tile_size_spin_box = self.findChild(QSpinBox, 'srmdNcnnVulkanTileSizeSpinBox')
-        self.srmd_ncnn_vulkan_model_combo_box = self.findChild(QComboBox, 'srmdNcnnVulkanModelComboBox')
-        self.srmd_ncnn_vulkan_gpu_id_spin_box = self.findChild(QSpinBox, 'srmdNcnnVulkanGpuIdSpinBox')
-        self.srmd_ncnn_vulkan_jobs_line_edit = self.findChild(QLineEdit, 'srmdNcnnVulkanJobsLineEdit')
-        self.srmd_ncnn_vulkan_tta_check_box = self.findChild(QCheckBox, 'srmdNcnnVulkanTtaCheckBox')
-
-        # anime4k
-        self.anime4kcpp_path_line_edit = self.findChild(QLineEdit, 'anime4kCppPathLineEdit')
-        self.enable_line_edit_file_drop(self.anime4kcpp_path_line_edit)
-        self.anime4kcpp_path_select_button = self.findChild(QPushButton, 'anime4kCppPathSelectButton')
-        self.anime4kcpp_path_select_button.clicked.connect(lambda: self.select_driver_binary_path(self.anime4kcpp_path_line_edit))
-        self.anime4kcpp_passes_spin_box = self.findChild(QSpinBox, 'anime4kCppPassesSpinBox')
-        self.anime4kcpp_push_color_count_spin_box = self.findChild(QSpinBox, 'anime4kCppPushColorCountSpinBox')
-        self.anime4kcpp_strength_color_spin_box = self.findChild(QDoubleSpinBox, 'anime4kCppStrengthColorSpinBox')
-        self.anime4kcpp_strength_gradient_spin_box = self.findChild(QDoubleSpinBox, 'anime4kCppStrengthGradientSpinBox')
-        self.anime4kcpp_threads_spin_box = self.findChild(QSpinBox, 'anime4kCppThreadsSpinBox')
-        self.anime4kcpp_pre_filters_spin_box = self.findChild(QSpinBox, 'anime4kCppPreFiltersSpinBox')
-        self.anime4kcpp_post_filters_spin_box = self.findChild(QSpinBox, 'anime4kCppPostFiltersSpinBox')
-        self.anime4kcpp_platform_id_spin_box = self.findChild(QSpinBox, 'anime4kCppPlatformIdSpinBox')
-        self.anime4kcpp_device_id_spin_box = self.findChild(QSpinBox, 'anime4kCppDeviceIdSpinBox')
-        self.anime4kcpp_codec_combo_box = self.findChild(QComboBox, 'anime4kCppCodecComboBox')
-        self.anime4kcpp_fast_mode_check_box = self.findChild(QCheckBox, 'anime4kCppFastModeCheckBox')
-        self.anime4kcpp_pre_processing_check_box = self.findChild(QCheckBox, 'anime4kCppPreProcessingCheckBox')
-        self.anime4kcpp_post_processing_check_box = self.findChild(QCheckBox, 'anime4kCppPostProcessingCheckBox')
-        self.anime4kcpp_gpu_mode_check_box = self.findChild(QCheckBox, 'anime4kCppGpuModeCheckBox')
-
-        # load configurations
-        self.load_configurations()
-
-    def dragEnterEvent(self, event):
-        if event.mimeData().hasUrls():
-            event.accept()
-        else:
-            event.ignore()
-
-    def dropEvent(self, event):
-        input_paths = [pathlib.Path(u.toLocalFile()) for u in event.mimeData().urls()]
-        for path in input_paths:
-            if (path.is_file() or path.is_dir()) and not self.input_table_path_exists(path):
-                self.input_table_data.append(path)
-
-        self.update_input_table()
-        self.update_output_path()
-
-    def enable_line_edit_file_drop(self, line_edit: QLineEdit):
-        line_edit.dragEnterEvent = self.dragEnterEvent
-        line_edit.dropEvent = lambda event: line_edit.setText(str(pathlib.Path(event.mimeData().urls()[0].toLocalFile()).absolute()))
-
-    @staticmethod
-    def read_config(config_file: pathlib.Path) -> dict:
-        """ read video2x configurations from config file
-
-        Arguments:
-            config_file {pathlib.Path} -- video2x configuration file pathlib.Path
-
-        Returns:
-            dict -- dictionary of video2x configuration
-        """
-
-        with open(config_file, 'r') as config:
-            return yaml.load(config, Loader=yaml.FullLoader)
-
-    def load_configurations(self):
-
-        # get config file path from line edit
-        config_file_path = pathlib.Path(os.path.expandvars(self.config_line_edit.text()))
-
-        # if file doesn't exist, return
-        if not config_file_path.is_file():
-            QErrorMessage(self).showMessage('Video2X configuration file not found, please specify manually.')
-            return
-
-        # read configuration dict from config file
-        self.config = self.read_config(config_file_path)
-
-        # load FFmpeg settings
-        self.ffmpeg_settings = self.config['ffmpeg']
-        self.ffmpeg_settings['ffmpeg_path'] = str(pathlib.Path(os.path.expandvars(self.ffmpeg_settings['ffmpeg_path'])).absolute())
-
-        # set cache directory path
-        if self.config['video2x']['video2x_cache_directory'] is None:
-            self.config['video2x']['video2x_cache_directory'] = str((pathlib.Path(tempfile.gettempdir()) / 'video2x').absolute())
-        self.cache_line_edit.setText(self.config['video2x']['video2x_cache_directory'])
-
-        # load preserve frames settings
-        self.preserve_frames_check_box.setChecked(self.config['video2x']['preserve_frames'])
-        self.start_button.setEnabled(True)
-
-        # waifu2x-caffe
-        settings = self.config['waifu2x_caffe']
-        self.waifu2x_caffe_scale_width_spin_box.setValue(settings['scale_width'])
-        self.waifu2x_caffe_scale_height_spin_box.setValue(settings['scale_height'])
-        self.waifu2x_caffe_path_line_edit.setText(str(pathlib.Path(os.path.expandvars(settings['path'])).absolute()))
-        self.waifu2x_caffe_mode_combo_box.setCurrentText(settings['mode'])
-        self.waifu2x_caffe_noise_level_spin_box.setValue(settings['noise_level'])
-        self.waifu2x_caffe_process_combo_box.setCurrentText(settings['process'])
-        self.waifu2x_caffe_crop_size_spin_box.setValue(settings['crop_size'])
-        self.waifu2x_caffe_output_quality_spin_box.setValue(settings['output_quality'])
-        self.waifu2x_caffe_output_depth_spin_box.setValue(settings['output_depth'])
-        self.waifu2x_caffe_batch_size_spin_box.setValue(settings['batch_size'])
-        self.waifu2x_caffe_gpu_spin_box.setValue(settings['gpu'])
-        self.waifu2x_caffe_tta_check_box.setChecked(bool(settings['tta']))
-
-        # waifu2x-converter-cpp
-        settings = self.config['waifu2x_converter_cpp']
-        self.waifu2x_converter_cpp_path_line_edit.setText(str(pathlib.Path(os.path.expandvars(settings['path'])).absolute()))
-        self.waifu2x_converter_cpp_png_compression_spin_box.setValue(settings['png-compression'])
-        self.waifu2x_converter_cpp_image_quality_spin_box.setValue(settings['image-quality'])
-        self.waifu2x_converter_cpp_block_size_spin_box.setValue(settings['block-size'])
-        self.waifu2x_converter_cpp_processor_spin_box.setValue(settings['processor'])
-        self.waifu2x_converter_cpp_noise_level_spin_box.setValue(settings['noise-level'])
-        self.waifu2x_converter_cpp_mode_combo_box.setCurrentText(settings['mode'])
-        self.waifu2x_converter_cpp_log_level_spin_box.setValue(settings['log-level'])
-        self.waifu2x_converter_cpp_disable_gpu_check_box.setChecked(settings['disable-gpu'])
-        self.waifu2x_converter_cpp_force_opencl_check_box.setChecked(settings['force-OpenCL'])
-        self.waifu2x_converter_cpp_tta_check_box.setChecked(bool(settings['tta']))
-
-        # waifu2x-ncnn-vulkan
-        settings = self.config['waifu2x_ncnn_vulkan']
-        self.waifu2x_ncnn_vulkan_path_line_edit.setText(str(pathlib.Path(os.path.expandvars(settings['path'])).absolute()))
-        self.waifu2x_ncnn_vulkan_noise_level_spin_box.setValue(settings['n'])
-        self.waifu2x_ncnn_vulkan_tile_size_spin_box.setValue(settings['t'])
-        self.waifu2x_ncnn_vulkan_gpu_id_spin_box.setValue(settings['g'])
-        self.waifu2x_ncnn_vulkan_jobs_line_edit.setText(settings['j'])
-        self.waifu2x_ncnn_vulkan_tta_check_box.setChecked(settings['x'])
-
-        # srmd-ncnn-vulkan
-        settings = self.config['srmd_ncnn_vulkan']
-        self.srmd_ncnn_vulkan_path_line_edit.setText(str(pathlib.Path(os.path.expandvars(settings['path'])).absolute()))
-        self.srmd_ncnn_vulkan_noise_level_spin_box.setValue(settings['n'])
-        self.srmd_ncnn_vulkan_tile_size_spin_box.setValue(settings['t'])
-        self.srmd_ncnn_vulkan_gpu_id_spin_box.setValue(settings['g'])
-        self.srmd_ncnn_vulkan_jobs_line_edit.setText(settings['j'])
-        self.srmd_ncnn_vulkan_tta_check_box.setChecked(settings['x'])
-
-        # anime4k
-        settings = self.config['anime4kcpp']
-        self.anime4kcpp_path_line_edit.setText(str(pathlib.Path(os.path.expandvars(settings['path'])).absolute()))
-        self.anime4kcpp_passes_spin_box.setValue(settings['passes'])
-        self.anime4kcpp_push_color_count_spin_box.setValue(settings['pushColorCount'])
-        self.anime4kcpp_strength_color_spin_box.setValue(settings['strengthColor'])
-        self.anime4kcpp_strength_gradient_spin_box.setValue(settings['strengthGradient'])
-        self.anime4kcpp_threads_spin_box.setValue(settings['threads'])
-        self.anime4kcpp_pre_filters_spin_box.setValue(settings['preFilters'])
-        self.anime4kcpp_post_filters_spin_box.setValue(settings['postFilters'])
-        self.anime4kcpp_platform_id_spin_box.setValue(settings['platformID'])
-        self.anime4kcpp_device_id_spin_box.setValue(settings['deviceID'])
-        self.anime4kcpp_codec_combo_box.setCurrentText(settings['codec'])
-        self.anime4kcpp_fast_mode_check_box.setChecked(settings['fastMode'])
-        self.anime4kcpp_pre_processing_check_box.setChecked(settings['preprocessing'])
-        self.anime4kcpp_post_processing_check_box.setChecked(settings['postprocessing'])
-        self.anime4kcpp_gpu_mode_check_box.setChecked(settings['GPUMode'])
-
-    def resolve_driver_settings(self):
-
-        # waifu2x-caffe
-        self.config['waifu2x_caffe']['scale_width'] = self.waifu2x_caffe_scale_width_spin_box.value()
-        self.config['waifu2x_caffe']['scale_height'] = self.waifu2x_caffe_scale_height_spin_box.value()
-        self.config['waifu2x_caffe']['path'] = os.path.expandvars(self.waifu2x_caffe_path_line_edit.text())
-        self.config['waifu2x_caffe']['mode'] = self.waifu2x_caffe_mode_combo_box.currentText()
-        self.config['waifu2x_caffe']['noise_level'] = self.waifu2x_caffe_noise_level_spin_box.value()
-        self.config['waifu2x_caffe']['process'] = self.waifu2x_caffe_process_combo_box.currentText()
-        self.config['waifu2x_caffe']['model_dir'] = str((pathlib.Path(self.config['waifu2x_caffe']['path']).parent / 'models' / self.waifu2x_caffe_model_combobox.currentText()).absolute())
-        self.config['waifu2x_caffe']['crop_size'] = self.waifu2x_caffe_crop_size_spin_box.value()
-        self.config['waifu2x_caffe']['output_quality'] = self.waifu2x_caffe_output_depth_spin_box.value()
-        self.config['waifu2x_caffe']['output_depth'] = self.waifu2x_caffe_output_depth_spin_box.value()
-        self.config['waifu2x_caffe']['batch_size'] = self.waifu2x_caffe_batch_size_spin_box.value()
-        self.config['waifu2x_caffe']['gpu'] = self.waifu2x_caffe_gpu_spin_box.value()
-        self.config['waifu2x_caffe']['tta'] = int(self.waifu2x_caffe_tta_check_box.isChecked())
-
-        # waifu2x-converter-cpp
-        self.config['waifu2x_converter_cpp']['path'] = os.path.expandvars(self.waifu2x_converter_cpp_path_line_edit.text())
-        self.config['waifu2x_converter_cpp']['png-compression'] = self.waifu2x_converter_cpp_png_compression_spin_box.value()
-        self.config['waifu2x_converter_cpp']['image-quality'] = self.waifu2x_converter_cpp_image_quality_spin_box.value()
-        self.config['waifu2x_converter_cpp']['block-size'] = self.waifu2x_converter_cpp_block_size_spin_box.value()
-        self.config['waifu2x_converter_cpp']['processor'] = self.waifu2x_converter_cpp_processor_spin_box.value()
-        self.config['waifu2x_converter_cpp']['model-dir'] = str((pathlib.Path(self.config['waifu2x_converter_cpp']['path']).parent / self.waifu2x_converter_cpp_model_combo_box.currentText()).absolute())
-        self.config['waifu2x_converter_cpp']['noise-level'] = self.waifu2x_converter_cpp_noise_level_spin_box.value()
-        self.config['waifu2x_converter_cpp']['mode'] = self.waifu2x_converter_cpp_mode_combo_box.currentText()
-        self.config['waifu2x_converter_cpp']['log-level'] = self.waifu2x_converter_cpp_log_level_spin_box.value()
-        self.config['waifu2x_converter_cpp']['disable-gpu'] = bool(self.waifu2x_converter_cpp_disable_gpu_check_box.isChecked())
-        self.config['waifu2x_converter_cpp']['force-OpenCL'] = bool(self.waifu2x_converter_cpp_force_opencl_check_box.isChecked())
-        self.config['waifu2x_converter_cpp']['tta'] = int(self.waifu2x_converter_cpp_tta_check_box.isChecked())
-
-        # waifu2x-ncnn-vulkan
-        self.config['waifu2x_ncnn_vulkan']['path'] = os.path.expandvars(self.waifu2x_ncnn_vulkan_path_line_edit.text())
-        self.config['waifu2x_ncnn_vulkan']['n'] = self.waifu2x_ncnn_vulkan_noise_level_spin_box.value()
-        self.config['waifu2x_ncnn_vulkan']['t'] = self.waifu2x_ncnn_vulkan_tile_size_spin_box.value()
-        self.config['waifu2x_ncnn_vulkan']['m'] = str((pathlib.Path(self.config['waifu2x_ncnn_vulkan']['path']).parent / self.waifu2x_ncnn_vulkan_model_combo_box.currentText()).absolute())
-        self.config['waifu2x_ncnn_vulkan']['g'] = self.waifu2x_ncnn_vulkan_gpu_id_spin_box.value()
-        self.config['waifu2x_ncnn_vulkan']['j'] = self.waifu2x_ncnn_vulkan_jobs_line_edit.text()
-        self.config['waifu2x_ncnn_vulkan']['x'] = self.waifu2x_ncnn_vulkan_tta_check_box.isChecked()
-
-        # srmd-ncnn-vulkan
-        self.config['srmd_ncnn_vulkan']['path'] = os.path.expandvars(self.srmd_ncnn_vulkan_path_line_edit.text())
-        self.config['srmd_ncnn_vulkan']['n'] = self.srmd_ncnn_vulkan_noise_level_spin_box.value()
-        self.config['srmd_ncnn_vulkan']['t'] = self.srmd_ncnn_vulkan_tile_size_spin_box.value()
-        self.config['srmd_ncnn_vulkan']['m'] = str((pathlib.Path(self.config['srmd_ncnn_vulkan']['path']).parent / self.srmd_ncnn_vulkan_model_combo_box.currentText()).absolute())
-        self.config['srmd_ncnn_vulkan']['g'] = self.srmd_ncnn_vulkan_gpu_id_spin_box.value()
-        self.config['srmd_ncnn_vulkan']['j'] = self.srmd_ncnn_vulkan_jobs_line_edit.text()
-        self.config['srmd_ncnn_vulkan']['x'] = self.srmd_ncnn_vulkan_tta_check_box.isChecked()
-
-        # anime4k
-        self.config['anime4kcpp']['path'] = os.path.expandvars(self.anime4kcpp_path_line_edit.text())
-        self.config['anime4kcpp']['passes'] = self.anime4kcpp_passes_spin_box.value()
-        self.config['anime4kcpp']['pushColorCount'] = self.anime4kcpp_push_color_count_spin_box.value()
-        self.config['anime4kcpp']['strengthColor'] = self.anime4kcpp_strength_color_spin_box.value()
-        self.config['anime4kcpp']['strengthGradient'] = self.anime4kcpp_strength_gradient_spin_box.value()
-        self.config['anime4kcpp']['threads'] = self.anime4kcpp_threads_spin_box.value()
-        self.config['anime4kcpp']['preFilters'] = self.anime4kcpp_pre_filters_spin_box.value()
-        self.config['anime4kcpp']['postFilters'] = self.anime4kcpp_post_filters_spin_box.value()
-        self.config['anime4kcpp']['platformID'] = self.anime4kcpp_platform_id_spin_box.value()
-        self.config['anime4kcpp']['deviceID'] = self.anime4kcpp_device_id_spin_box.value()
-        self.config['anime4kcpp']['codec'] = self.anime4kcpp_codec_combo_box.currentText()
-        self.config['anime4kcpp']['fastMode'] = bool(self.anime4kcpp_fast_mode_check_box.isChecked())
-        self.config['anime4kcpp']['preprocessing'] = bool(self.anime4kcpp_pre_processing_check_box.isChecked())
-        self.config['anime4kcpp']['postprocessing'] = bool(self.anime4kcpp_post_processing_check_box.isChecked())
-        self.config['anime4kcpp']['GPUMode'] = bool(self.anime4kcpp_gpu_mode_check_box.isChecked())
-
-    def update_gui_for_driver(self):
-        current_driver = AVAILABLE_DRIVERS[self.driver_combo_box.currentText()]
-
-        # update scale ratio constraints
-        if current_driver in ['waifu2x_caffe', 'waifu2x_converter_cpp', 'anime4kcpp']:
-            self.scale_ratio_double_spin_box.setMinimum(0.0)
-            self.scale_ratio_double_spin_box.setMaximum(999.0)
-            self.scale_ratio_double_spin_box.setValue(2.0)
-        elif current_driver == 'waifu2x_ncnn_vulkan':
-            self.scale_ratio_double_spin_box.setMinimum(1.0)
-            self.scale_ratio_double_spin_box.setMaximum(2.0)
-            self.scale_ratio_double_spin_box.setValue(2.0)
-        elif current_driver == 'srmd_ncnn_vulkan':
-            self.scale_ratio_double_spin_box.setMinimum(2.0)
-            self.scale_ratio_double_spin_box.setMaximum(4.0)
-            self.scale_ratio_double_spin_box.setValue(2.0)
-
-        # update preferred processes/threads count
-        if current_driver == 'anime4kcpp':
-            self.processes_spin_box.setValue(16)
-        else:
-            self.processes_spin_box.setValue(1)
-
-    def update_input_table(self):
-        # HACK: use insertRow, removeRow and signals
-        del self.input_table_model
-        self.input_table_model = InputTableModel(self.input_table_data)
-        self.input_table_view.setModel(self.input_table_model)
-
-    def input_table_delete_selected(self):
-        items_to_delete = []
-        for index in [i.row() for i in self.input_table_view.selectedIndexes()]:
-            items_to_delete.append(self.input_table_data[index])
-
-        for item in items_to_delete:
-            self.input_table_data.remove(item)
-        self.update_input_table()
-
-    def input_table_clear_all(self):
-        self.input_table_data = []
-        self.update_input_table()
-
-    def input_table_path_exists(self, input_path):
-        for path in self.input_table_data:
-            # not using Path.samefile since file may not exist
-            if str(path.absolute()) == str(input_path.absolute()):
-                return True
-        return False
-
-    def select_file(self, *args, **kwargs) -> pathlib.Path:
-        file_selected = QFileDialog.getOpenFileName(self, *args, **kwargs)
-        if not isinstance(file_selected, tuple) or file_selected[0] == '':
-            return None
-        return pathlib.Path(file_selected[0])
-
-    def select_folder(self, *args, **kwargs) -> pathlib.Path:
-        folder_selected = QFileDialog.getExistingDirectory(self, *args, **kwargs)
-        if folder_selected == '':
-            return None
-        return pathlib.Path(folder_selected)
-
-    def update_output_path(self):
-        # if there is more than one input
-        if len(self.input_table_data) != 1:
-            return
-
-        input_path = self.input_table_data[0]
-        # give up if input path doesn't exist or isn't a file or a directory
-        if not input_path.exists() or not (input_path.is_file() or input_path.is_dir()):
-            return
-
-        if input_path.is_file():
-            output_path = input_path.parent / f'{input_path.stem}_output.mp4'
-        elif input_path.is_dir():
-            output_path = input_path.parent / f'{input_path.stem}_output'
-
-        # try up to 1000 times
-        output_path_id = 0
-        while output_path.exists() and output_path_id <= 1000:
-            if input_path.is_file():
-                output_path = input_path.parent / pathlib.Path(f'{input_path.stem}_output_{output_path_id}.mp4')
-            elif input_path.is_dir():
-                output_path = input_path.parent / pathlib.Path(f'{input_path.stem}_output_{output_path_id}')
-            output_path_id += 1
-
-        if not output_path.exists():
-            self.output_line_edit.setText(str(output_path.absolute()))
-
-    def select_input_file(self):
-        if ((input_file := self.select_file('Select Input File')) is None or
-                self.input_table_path_exists(input_file)):
-            return
-        self.input_table_data.append(input_file)
-        self.update_output_path()
-        self.update_input_table()
-
-    def select_input_folder(self):
-        if ((input_folder := self.select_folder('Select Input Folder')) is None or
-                self.input_table_path_exists(input_folder)):
-            return
-        self.input_table_data.append(input_folder)
-        self.update_output_path()
-        self.update_input_table()
-
-    def select_output_file(self):
-        if (output_file := self.select_file('Select Output File')) is None:
-            return
-        self.output_line_edit.setText(str(output_file.absolute()))
-
-    def select_output_folder(self):
-        if (output_folder := self.select_folder('Select Output Folder')) is None:
-            return
-        self.output_line_edit.setText(str(output_folder.absolute()))
-
-    def select_cache_folder(self):
-        if (cache_folder := self.select_folder('Select Cache Folder')) is None:
-            return
-        self.cache_line_edit.setText(str(cache_folder.absolute()))
-
-    def select_config_file(self):
-        if (config_file := self.select_file('Select Config File', filter='(YAML files (*.yaml))')) is None:
-            return
-        self.config_line_edit.setText(str(config_file.absolute()))
-        self.load_configurations()
-
-    def select_driver_binary_path(self, driver_line_edit):
-        if (driver_binary_path := self.select_file('Select Driver Binary File')) is None:
-            return
-        driver_line_edit.setText(str(driver_binary_path.absolute()))
-
-    def show_about(self, message: str):
-        message_box = QMessageBox(self)
-        message_box.setWindowTitle('About Video2X')
-        message_box.setIconPixmap(QtGui.QPixmap(self.video2x_icon_path).scaled(64, 64))
-        message_box.setTextFormat(Qt.MarkdownText)
-        message_box.setText(LEGAL_INFO)
-        message_box.exec_()
-
-    def show_information(self, message: str):
-        message_box = QMessageBox(self)
-        message_box.setWindowTitle('Information')
-        message_box.setIcon(QMessageBox.Information)
-        message_box.setText(message)
-        message_box.exec_()
-
-    def show_warning(self, message: str):
-        message_box = QMessageBox(self)
-        message_box.setWindowTitle('Warning')
-        message_box.setIcon(QMessageBox.Warning)
-        message_box.setText(message)
-        message_box.exec_()
-
-    def show_error(self, exception: Exception):
-        # QErrorMessage(self).showMessage(message.replace('\n', '<br>'))
-        message_box = QMessageBox(self)
-        message_box.setWindowTitle('Error')
-        message_box.setIcon(QMessageBox.Critical)
-        message_box.setTextFormat(Qt.MarkdownText)
-
-        error_message = '''Upscaler ran into an error:\\
-**{}**\\
-Check the console output for details.\\
-When reporting an error, please include console output.\\
-You can [submit an issue on GitHub](https://github.com/k4yt3x/video2x/issues/new?assignees=K4YT3X&labels=bug&template=bug-report.md&title={}) to report this error.'''
-        message_box.setText(error_message.format(exception, urllib.parse.quote(str(exception))))
-        message_box.exec_()
-
-    def progress_monitor(self, progress_callback):
-
-        # initialize progress bar values
-        upscale_begin_time = time.time()
-        progress_callback.emit((upscale_begin_time, 0, 0, 0, 0, pathlib.Path(), pathlib.Path()))
-
-        # keep querying upscaling process and feed information to callback signal
-        while self.upscaler.running:
-
-            progress_callback.emit((upscale_begin_time,
-                                    self.upscaler.total_frames_upscaled,
-                                    self.upscaler.total_frames,
-                                    self.upscaler.total_processed,
-                                    self.upscaler.total_videos,
-                                    self.upscaler.current_input_video,
-                                    self.upscaler.last_frame_upscaled))
-            time.sleep(1)
-
-        # upscale process will stop at 99%
-        # so it's set to 100 manually when all is done
-        progress_callback.emit((upscale_begin_time, 0, 0, 0, 0, pathlib.Path(), pathlib.Path()))
-
-    def set_progress(self, progress_information: tuple):
-        upscale_begin_time = progress_information[0]
-        total_frames_upscaled = progress_information[1]
-        total_frames = progress_information[2]
-        total_processed = progress_information[3]
-        total_videos = progress_information[4]
-        current_input_video = progress_information[5]
-        last_frame_upscaled = progress_information[6]
-
-        # calculate fields based on frames and time elapsed
-        time_elapsed = time.time() - upscale_begin_time
-        try:
-            rate = total_frames_upscaled / (time.time() - upscale_begin_time)
-            time_remaining = (total_frames - total_frames_upscaled) / rate
-        except Exception:
-            rate = 0.0
-            time_remaining = 0.0
-
-        # set calculated values in GUI
-        self.current_progress_bar.setMaximum(total_frames)
-        self.current_progress_bar.setValue(total_frames_upscaled)
-        self.frames_label.setText('Frames: {}/{}'.format(total_frames_upscaled, total_frames))
-        self.time_elapsed_label.setText('Time Elapsed: {}'.format(time.strftime("%H:%M:%S", time.gmtime(time_elapsed))))
-        self.time_remaining_label.setText('Time Remaining: {}'.format(time.strftime("%H:%M:%S", time.gmtime(time_remaining))))
-        self.rate_label.setText('Rate (FPS): {}'.format(round(rate, 2)))
-        self.overall_progress_label.setText('Overall Progress: {}/{}'.format(total_processed, total_videos))
-        self.overall_progress_bar.setMaximum(total_videos)
-        self.overall_progress_bar.setValue(total_processed)
-        self.currently_processing_label.setText('Currently Processing: {}'.format(str(current_input_video.name)))
-
-        # if show frame is checked, show preview image
-        if self.frame_preview_show_preview_check_box.isChecked() and last_frame_upscaled.is_file():
-            last_frame_pixmap = QtGui.QPixmap(str(last_frame_upscaled.absolute()))
-            # the -2 here behind geometry subtracts frame size from width and height
-            self.frame_preview_label.setPixmap(last_frame_pixmap.scaled(self.frame_preview_label.width() - 2,
-                                                                        self.frame_preview_label.height() - 2,
-                                                                        Qt.KeepAspectRatio))
-
-            # if keep aspect ratio is checked, don't stretch image
-            if self.frame_preview_keep_aspect_ratio_check_box.isChecked():
-                self.frame_preview_label.setScaledContents(False)
-            else:
-                self.frame_preview_label.setScaledContents(True)
-
-            # display image in label
-            self.frame_preview_label.show()
-
-        # if show frame is unchecked, clear image
-        elif self.frame_preview_show_preview_check_box.isChecked() is False:
-            self.frame_preview_label.clear()
-
-    def reset_progress_display(self):
-        # reset progress display UI elements
-        self.current_progress_bar.setMaximum(100)
-        self.current_progress_bar.setValue(0)
-        self.frames_label.setText('Frames: {}/{}'.format(0, 0))
-        self.time_elapsed_label.setText('Time Elapsed: {}'.format(time.strftime("%H:%M:%S", time.gmtime(0))))
-        self.time_remaining_label.setText('Time Remaining: {}'.format(time.strftime("%H:%M:%S", time.gmtime(0))))
-        self.rate_label.setText('Rate (FPS): {}'.format(0.0))
-        self.overall_progress_label.setText('Overall Progress: {}/{}'.format(0, 0))
-        self.overall_progress_bar.setMaximum(100)
-        self.overall_progress_bar.setValue(0)
-        self.currently_processing_label.setText('Currently Processing:')
-
-    def start(self):
-
-        # start execution
-        try:
-            # start timer
-            self.begin_time = time.time()
-
-            # resolve input and output directories from GUI
-            if len(self.input_table_data) == 0:
-                self.show_warning('Input path unspecified')
-                return
-            if self.output_line_edit.text().strip() == '':
-                self.show_warning('Output path unspecified')
-                return
-
-            if len(self.input_table_data) == 1:
-                input_directory = self.input_table_data[0]
-            else:
-                input_directory = self.input_table_data
-
-            # resolve output directory
-            output_directory = pathlib.Path(os.path.expandvars(self.output_line_edit.text()))
-
-            # load driver settings from GUI
-            self.resolve_driver_settings()
-
-            # load driver settings for the current driver
-            self.driver_settings = self.config[AVAILABLE_DRIVERS[self.driver_combo_box.currentText()]]
-
-            self.upscaler = Upscaler(input_path=input_directory,
-                                     output_path=output_directory,
-                                     driver_settings=self.driver_settings,
-                                     ffmpeg_settings=self.ffmpeg_settings)
-
-            # set optional options
-            self.upscaler.driver = AVAILABLE_DRIVERS[self.driver_combo_box.currentText()]
-            self.upscaler.scale_ratio = self.scale_ratio_double_spin_box.value()
-            self.upscaler.processes = self.processes_spin_box.value()
-            self.upscaler.video2x_cache_directory = pathlib.Path(os.path.expandvars(self.cache_line_edit.text()))
-            self.upscaler.image_format = self.config['video2x']['image_format'].lower()
-            self.upscaler.preserve_frames = bool(self.preserve_frames_check_box.isChecked())
-
-            # run upscaler
-            worker = UpscalerWorker(self.upscaler.run)
-            worker.signals.error.connect(self.upscale_errored)
-            worker.signals.interrupted.connect(self.upscale_interrupted)
-            worker.signals.finished.connect(self.upscale_successful)
-            self.threadpool.start(worker)
-
-            # start progress monitoring
-            progress_bar_worker = ProgressMonitorWorkder(self.progress_monitor)
-            progress_bar_worker.signals.progress.connect(self.set_progress)
-            self.threadpool.start(progress_bar_worker)
-
-            self.start_button.setEnabled(False)
-            self.stop_button.setEnabled(True)
-
-        except Exception as e:
-            traceback.print_exc()
-            self.upscale_errored(e)
-
-    def upscale_errored(self, exception: Exception):
-        self.show_error(exception)
-        self.threadpool.waitForDone(5)
-        self.start_button.setEnabled(True)
-        self.stop_button.setEnabled(False)
-        self.reset_progress_display()
-
-    def upscale_interrupted(self):
-        self.show_information('Upscale has been interrupted')
-        self.threadpool.waitForDone(5)
-        self.start_button.setEnabled(True)
-        self.stop_button.setEnabled(False)
-        self.reset_progress_display()
-
-    def upscale_successful(self):
-        # if all threads have finished
-        self.threadpool.waitForDone(5)
-        self.show_information('Upscale finished successfully, taking {} seconds'.format(round((time.time() - self.begin_time), 5)))
-        self.start_button.setEnabled(True)
-        self.stop_button.setEnabled(False)
-        self.reset_progress_display()
-
-    def stop(self):
-        with contextlib.suppress(AttributeError):
-            self.upscaler.running = False
-
-    def closeEvent(self, event):
-        # try cleaning up temp directories
-        self.stop()
-        event.accept()
-
-
-# this file shouldn't be imported
-if __name__ == '__main__':
-    try:
-        app = QApplication(sys.argv)
-        window = Video2XMainWindow()
-        window.show()
-        app.exec_()
-
-    # on GUI exception, print error message in console
-    # and hold window open using input()
-    except Exception:
-        traceback.print_exc()
-        input('Press enter to close')

src/video2x_gui.pyproject

@@ -1,3 +0,0 @@
-{
-    "files": ["video2x_gui.ui","video2x_gui.py"]
-}

src/video2x_gui.pyproject.user

@@ -1,190 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-<!DOCTYPE QtCreatorProject>
-<!-- Written by QtCreator 4.12.0, 2020-05-11T04:13:22. -->
-<qtcreator>
- <data>
-  <variable>EnvironmentId</variable>
-  <value type="QByteArray">{25c1e4a4-6b4a-4828-883b-6f32c3f7eba0}</value>
- </data>
- <data>
-  <variable>ProjectExplorer.Project.ActiveTarget</variable>
-  <value type="int">0</value>
- </data>
- <data>
-  <variable>ProjectExplorer.Project.EditorSettings</variable>
-  <valuemap type="QVariantMap">
-   <value type="bool" key="EditorConfiguration.AutoIndent">true</value>
-   <value type="bool" key="EditorConfiguration.AutoSpacesForTabs">false</value>
-   <value type="bool" key="EditorConfiguration.CamelCaseNavigation">true</value>
-   <valuemap type="QVariantMap" key="EditorConfiguration.CodeStyle.0">
-    <value type="QString" key="language">Cpp</value>
-    <valuemap type="QVariantMap" key="value">
-     <value type="QByteArray" key="CurrentPreferences">CppGlobal</value>
-    </valuemap>
-   </valuemap>
-   <valuemap type="QVariantMap" key="EditorConfiguration.CodeStyle.1">
-    <value type="QString" key="language">QmlJS</value>
-    <valuemap type="QVariantMap" key="value">
-     <value type="QByteArray" key="CurrentPreferences">QmlJSGlobal</value>
-    </valuemap>
-   </valuemap>
-   <value type="int" key="EditorConfiguration.CodeStyle.Count">2</value>
-   <value type="QByteArray" key="EditorConfiguration.Codec">UTF-8</value>
-   <value type="bool" key="EditorConfiguration.ConstrainTooltips">false</value>
-   <value type="int" key="EditorConfiguration.IndentSize">4</value>
-   <value type="bool" key="EditorConfiguration.KeyboardTooltips">false</value>
-   <value type="int" key="EditorConfiguration.MarginColumn">80</value>
-   <value type="bool" key="EditorConfiguration.MouseHiding">true</value>
-   <value type="bool" key="EditorConfiguration.MouseNavigation">true</value>
-   <value type="int" key="EditorConfiguration.PaddingMode">1</value>
-   <value type="bool" key="EditorConfiguration.ScrollWheelZooming">true</value>
-   <value type="bool" key="EditorConfiguration.ShowMargin">false</value>
-   <value type="int" key="EditorConfiguration.SmartBackspaceBehavior">0</value>
-   <value type="bool" key="EditorConfiguration.SmartSelectionChanging">true</value>
-   <value type="bool" key="EditorConfiguration.SpacesForTabs">true</value>
-   <value type="int" key="EditorConfiguration.TabKeyBehavior">0</value>
-   <value type="int" key="EditorConfiguration.TabSize">8</value>
-   <value type="bool" key="EditorConfiguration.UseGlobal">true</value>
-   <value type="int" key="EditorConfiguration.Utf8BomBehavior">1</value>
-   <value type="bool" key="EditorConfiguration.addFinalNewLine">true</value>
-   <value type="bool" key="EditorConfiguration.cleanIndentation">true</value>
-   <value type="bool" key="EditorConfiguration.cleanWhitespace">true</value>
-   <value type="bool" key="EditorConfiguration.inEntireDocument">false</value>
-  </valuemap>
- </data>
- <data>
-  <variable>ProjectExplorer.Project.PluginSettings</variable>
-  <valuemap type="QVariantMap">
-   <valuelist type="QVariantList" key="ClangCodeModel.CustomCommandLineKey">
-    <value type="QString">-fno-delayed-template-parsing</value>
-   </valuelist>
-   <value type="bool" key="ClangCodeModel.UseGlobalConfig">true</value>
-   <value type="QString" key="ClangCodeModel.WarningConfigId">Builtin.Questionable</value>
-   <valuemap type="QVariantMap" key="ClangTools">
-    <value type="bool" key="ClangTools.BuildBeforeAnalysis">true</value>
-    <value type="QString" key="ClangTools.DiagnosticConfig">Builtin.DefaultTidyAndClazy</value>
-    <value type="int" key="ClangTools.ParallelJobs">8</value>
-    <valuelist type="QVariantList" key="ClangTools.SelectedDirs"/>
-    <valuelist type="QVariantList" key="ClangTools.SelectedFiles"/>
-    <valuelist type="QVariantList" key="ClangTools.SuppressedDiagnostics"/>
-    <value type="bool" key="ClangTools.UseGlobalSettings">true</value>
-   </valuemap>
-  </valuemap>
- </data>
- <data>
-  <variable>ProjectExplorer.Project.Target.0</variable>
-  <valuemap type="QVariantMap">
-   <value type="QString" key="ProjectExplorer.ProjectConfiguration.DefaultDisplayName">Desktop (x86-windows-msvc2019-pe-64bit)</value>
-   <value type="QString" key="ProjectExplorer.ProjectConfiguration.DisplayName">Desktop (x86-windows-msvc2019-pe-64bit)</value>
-   <value type="QString" key="ProjectExplorer.ProjectConfiguration.Id">{5450e32e-f498-4c15-a0a5-7456f3375388}</value>
-   <value type="int" key="ProjectExplorer.Target.ActiveBuildConfiguration">-1</value>
-   <value type="int" key="ProjectExplorer.Target.ActiveDeployConfiguration">0</value>
-   <value type="int" key="ProjectExplorer.Target.ActiveRunConfiguration">0</value>
-   <value type="int" key="ProjectExplorer.Target.BuildConfigurationCount">0</value>
-   <valuemap type="QVariantMap" key="ProjectExplorer.Target.DeployConfiguration.0">
-    <valuemap type="QVariantMap" key="ProjectExplorer.BuildConfiguration.BuildStepList.0">
-     <value type="int" key="ProjectExplorer.BuildStepList.StepsCount">0</value>
-     <value type="QString" key="ProjectExplorer.ProjectConfiguration.DefaultDisplayName">Deploy</value>
-     <value type="QString" key="ProjectExplorer.ProjectConfiguration.DisplayName">Deploy</value>
-     <value type="QString" key="ProjectExplorer.ProjectConfiguration.Id">ProjectExplorer.BuildSteps.Deploy</value>
-    </valuemap>
-    <value type="int" key="ProjectExplorer.BuildConfiguration.BuildStepListCount">1</value>
-    <valuemap type="QVariantMap" key="ProjectExplorer.DeployConfiguration.CustomData"/>
-    <value type="bool" key="ProjectExplorer.DeployConfiguration.CustomDataEnabled">false</value>
-    <value type="QString" key="ProjectExplorer.ProjectConfiguration.Id">ProjectExplorer.DefaultDeployConfiguration</value>
-   </valuemap>
-   <value type="int" key="ProjectExplorer.Target.DeployConfigurationCount">1</value>
-   <valuemap type="QVariantMap" key="ProjectExplorer.Target.PluginSettings"/>
-   <valuemap type="QVariantMap" key="ProjectExplorer.Target.RunConfiguration.0">
-    <value type="QString" key="Analyzer.Perf.CallgraphMode">dwarf</value>
-    <valuelist type="QVariantList" key="Analyzer.Perf.Events">
-     <value type="QString">cpu-cycles</value>
-    </valuelist>
-    <valuelist type="QVariantList" key="Analyzer.Perf.ExtraArguments"/>
-    <value type="int" key="Analyzer.Perf.Frequency">250</value>
-    <valuelist type="QVariantList" key="Analyzer.Perf.RecordArguments">
-     <value type="QString">-e</value>
-     <value type="QString">cpu-cycles</value>
-     <value type="QString">--call-graph</value>
-     <value type="QString">dwarf,4096</value>
-     <value type="QString">-F</value>
-     <value type="QString">250</value>
-    </valuelist>
-    <value type="QString" key="Analyzer.Perf.SampleMode">-F</value>
-    <value type="bool" key="Analyzer.Perf.Settings.UseGlobalSettings">true</value>
-    <value type="int" key="Analyzer.Perf.StackSize">4096</value>
-    <value type="bool" key="Analyzer.QmlProfiler.AggregateTraces">false</value>
-    <value type="bool" key="Analyzer.QmlProfiler.FlushEnabled">false</value>
-    <value type="uint" key="Analyzer.QmlProfiler.FlushInterval">1000</value>
-    <value type="QString" key="Analyzer.QmlProfiler.LastTraceFile"></value>
-    <value type="bool" key="Analyzer.QmlProfiler.Settings.UseGlobalSettings">true</value>
-    <valuelist type="QVariantList" key="Analyzer.Valgrind.AddedSuppressionFiles"/>
-    <value type="bool" key="Analyzer.Valgrind.Callgrind.CollectBusEvents">false</value>
-    <value type="bool" key="Analyzer.Valgrind.Callgrind.CollectSystime">false</value>
-    <value type="bool" key="Analyzer.Valgrind.Callgrind.EnableBranchSim">false</value>
-    <value type="bool" key="Analyzer.Valgrind.Callgrind.EnableCacheSim">false</value>
-    <value type="bool" key="Analyzer.Valgrind.Callgrind.EnableEventToolTips">true</value>
-    <value type="double" key="Analyzer.Valgrind.Callgrind.MinimumCostRatio">0.01</value>
-    <value type="double" key="Analyzer.Valgrind.Callgrind.VisualisationMinimumCostRatio">10</value>
-    <value type="bool" key="Analyzer.Valgrind.FilterExternalIssues">true</value>
-    <value type="QString" key="Analyzer.Valgrind.KCachegrindExecutable">kcachegrind</value>
-    <value type="int" key="Analyzer.Valgrind.LeakCheckOnFinish">1</value>
-    <value type="int" key="Analyzer.Valgrind.NumCallers">25</value>
-    <valuelist type="QVariantList" key="Analyzer.Valgrind.RemovedSuppressionFiles"/>
-    <value type="int" key="Analyzer.Valgrind.SelfModifyingCodeDetection">1</value>
-    <value type="bool" key="Analyzer.Valgrind.Settings.UseGlobalSettings">true</value>
-    <value type="bool" key="Analyzer.Valgrind.ShowReachable">false</value>
-    <value type="bool" key="Analyzer.Valgrind.TrackOrigins">true</value>
-    <value type="QString" key="Analyzer.Valgrind.ValgrindExecutable">valgrind</value>
-    <valuelist type="QVariantList" key="Analyzer.Valgrind.VisibleErrorKinds">
-     <value type="int">0</value>
-     <value type="int">1</value>
-     <value type="int">2</value>
-     <value type="int">3</value>
-     <value type="int">4</value>
-     <value type="int">5</value>
-     <value type="int">6</value>
-     <value type="int">7</value>
-     <value type="int">8</value>
-     <value type="int">9</value>
-     <value type="int">10</value>
-     <value type="int">11</value>
-     <value type="int">12</value>
-     <value type="int">13</value>
-     <value type="int">14</value>
-    </valuelist>
-    <value type="int" key="PE.EnvironmentAspect.Base">2</value>
-    <valuelist type="QVariantList" key="PE.EnvironmentAspect.Changes"/>
-    <value type="QString" key="ProjectExplorer.ProjectConfiguration.DisplayName">video2x_gui</value>
-    <value type="QString" key="ProjectExplorer.ProjectConfiguration.Id">PythonEditor.RunConfiguration.C:/Users/k4yt3x/Documents/Projects/video2x/src/video2x_gui.py</value>
-    <value type="QString" key="ProjectExplorer.RunConfiguration.BuildKey">C:/Users/k4yt3x/Documents/Projects/video2x/src/video2x_gui.py</value>
-    <value type="bool" key="PythonEditor.RunConfiguation.Buffered">false</value>
-    <value type="QString" key="PythonEditor.RunConfiguation.Interpreter">{ffc39ed5-fbe0-4af8-8b16-9a60c393bdb8}</value>
-    <value type="QString" key="PythonEditor.RunConfiguation.Script">C:\Users\k4yt3x\Documents\Projects\video2x\src\video2x_gui.py</value>
-    <value type="QString" key="RunConfiguration.Arguments"></value>
-    <value type="bool" key="RunConfiguration.Arguments.multi">false</value>
-    <value type="QString" key="RunConfiguration.OverrideDebuggerStartup"></value>
-    <value type="bool" key="RunConfiguration.UseCppDebugger">false</value>
-    <value type="bool" key="RunConfiguration.UseCppDebuggerAuto">true</value>
-    <value type="bool" key="RunConfiguration.UseMultiProcess">false</value>
-    <value type="bool" key="RunConfiguration.UseQmlDebugger">false</value>
-    <value type="bool" key="RunConfiguration.UseQmlDebuggerAuto">true</value>
-    <value type="QString" key="RunConfiguration.WorkingDirectory"></value>
-    <value type="QString" key="RunConfiguration.WorkingDirectory.default"></value>
-   </valuemap>
-   <value type="int" key="ProjectExplorer.Target.RunConfigurationCount">1</value>
-  </valuemap>
- </data>
- <data>
-  <variable>ProjectExplorer.Project.TargetCount</variable>
-  <value type="int">1</value>
- </data>
- <data>
-  <variable>ProjectExplorer.Project.Updater.FileVersion</variable>
-  <value type="int">22</value>
- </data>
- <data>
-  <variable>Version</variable>
-  <value type="int">22</value>
- </data>
-</qtcreator>

src/video2x_setup.py

@@ -1,412 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Name: Video2X Setup Script
-Creator: K4YT3X
-Date Created: November 28, 2018
-Last Modified: May 8, 2020
-
-Editor: BrianPetkovsek
-Editor: SAT3LL
-
-Description: This script helps installing all dependencies of video2x
-and generates a configuration for it.
-
-Installation Details:
-- ffmpeg: %LOCALAPPDATA%\\video2x\\ffmpeg
-- waifu2x-caffe: %LOCALAPPDATA%\\video2x\\waifu2x-caffe
-- waifu2x-cpp-converter: %LOCALAPPDATA%\\video2x\\waifu2x-converter-cpp
-- waifu2x_ncnn_vulkan: %LOCALAPPDATA%\\video2x\\waifu2x-ncnn-vulkan
-- anime4kcpp: %LOCALAPPDATA%\\video2x\\anime4kcpp
-- srmd_ncnn_vulkan: %LOCALAPPDATA%\\video2x\\srmd-ncnn-vulkan
-"""
-
-# built-in imports
-from datetime import timedelta
-import argparse
-import contextlib
-import os
-import pathlib
-import re
-import shutil
-import subprocess
-import sys
-import tempfile
-import time
-import traceback
-import urllib
-import zipfile
-
-# Requests doesn't come with windows, therefore
-# it will be installed as a dependency and imported
-# later in the script.
-# import requests
-
-VERSION = '1.8.0'
-
-# global static variables
-LOCALAPPDATA = pathlib.Path(os.getenv('localappdata'))
-VIDEO2X_CONFIG = pathlib.Path(__file__).parent.absolute() / 'video2x.yaml'
-DRIVER_OPTIONS = ['all', 'ffmpeg', 'waifu2x_caffe', 'waifu2x_converter_cpp', 'waifu2x_ncnn_vulkan', 'anime4kcpp', 'srmd_ncnn_vulkan']
-
-
-def parse_arguments():
-    """Processes CLI arguments
-    """
-    parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter)
-
-    # video options
-    general_options = parser.add_argument_group('General Options')
-    general_options.add_argument('-d', '--driver', help='driver to download and configure', action='store', choices=DRIVER_OPTIONS, default='all')
-
-    # parse arguments
-    return parser.parse_args()
-
-
-class Video2xSetup:
-    """ Install dependencies for video2x video enlarger
-
-    This library is meant to be executed as a stand-alone
-    script. All files will be installed under %LOCALAPPDATA%\\video2x.
-    """
-
-    def __init__(self, driver, download_python_modules):
-        self.driver = driver
-        self.download_python_modules = download_python_modules
-        self.trash = []
-
-    def run(self):
-        if self.download_python_modules:
-            print('\nInstalling Python libraries')
-            self._install_python_requirements()
-
-        if self.driver == 'all':
-            self._install_ffmpeg()
-            self._install_waifu2x_caffe()
-            self._install_waifu2x_converter_cpp()
-            self._install_waifu2x_ncnn_vulkan()
-            self._install_anime4kcpp()
-            self._install_srmd_ncnn_vulkan()
-        elif self.driver == 'ffmpeg':
-            self._install_ffmpeg()
-        elif self.driver == 'waifu2x_caffe':
-            self._install_waifu2x_caffe()
-        elif self.driver == 'waifu2x_converter_cpp':
-            self._install_waifu2x_converter_cpp()
-        elif self.driver == 'waifu2x_ncnn_vulkan':
-            self._install_waifu2x_ncnn_vulkan()
-        elif self.driver == 'anime4kcpp':
-            self._install_anime4kcpp()
-        elif self.driver == 'srmd_ncnn_vulkan':
-            self._install_srmd_ncnn_vulkan()
-
-        print('\nGenerating Video2X configuration file')
-        self._generate_config()
-
-        print('\nCleaning up temporary files')
-        self._cleanup()
-
-    def _install_python_requirements(self):
-        """ Read requirements.txt and return its content
-        """
-        pip_install('requirements.txt')
-
-    def _cleanup(self):
-        """ Cleanup all the temp files downloaded
-        """
-        for file in self.trash:
-            try:
-                if file.is_dir():
-                    print(f'Deleting directory: {file}')
-                    shutil.rmtree(file)
-                else:
-                    print(f'Deleting file: {file}')
-                    file.unlink()
-            except Exception:
-                print(f'Error deleting: {file}')
-                traceback.print_exc()
-
-    def _install_ffmpeg(self):
-        """ Install FFMPEG
-        """
-        print('\nInstalling FFmpeg')
-
-        latest_release = 'https://ffmpeg.zeranoe.com/builds/win64/static/ffmpeg-latest-win64-static.zip'
-
-        ffmpeg_zip = download(latest_release, tempfile.gettempdir())
-        self.trash.append(ffmpeg_zip)
-
-        with zipfile.ZipFile(ffmpeg_zip) as zipf:
-            zipf.extractall(LOCALAPPDATA / 'video2x')
-
-    def _install_waifu2x_caffe(self):
-        """ Install waifu2x_caffe
-        """
-        print('\nInstalling waifu2x-caffe')
-        import requests
-
-        # Get latest release of waifu2x-caffe via GitHub API
-        latest_release = requests.get('https://api.github.com/repos/lltcggie/waifu2x-caffe/releases/latest').json()
-
-        for a in latest_release['assets']:
-            if 'waifu2x-caffe.zip' in a['browser_download_url']:
-                waifu2x_caffe_zip = download(a['browser_download_url'], tempfile.gettempdir())
-                self.trash.append(waifu2x_caffe_zip)
-
-        with zipfile.ZipFile(waifu2x_caffe_zip) as zipf:
-            zipf.extractall(LOCALAPPDATA / 'video2x')
-
-    def _install_waifu2x_converter_cpp(self):
-        """ Install waifu2x_caffe
-        """
-        print('\nInstalling waifu2x-converter-cpp')
-        import requests
-
-        # Get latest release of waifu2x-caffe via GitHub API
-        latest_release = requests.get('https://api.github.com/repos/DeadSix27/waifu2x-converter-cpp/releases/latest').json()
-
-        for a in latest_release['assets']:
-            if re.search(r'waifu2x-DeadSix27-win64_v[0-9]*\.zip', a['browser_download_url']):
-                waifu2x_converter_cpp_zip = download(a['browser_download_url'], tempfile.gettempdir())
-                self.trash.append(waifu2x_converter_cpp_zip)
-
-        with zipfile.ZipFile(waifu2x_converter_cpp_zip) as zipf:
-            zipf.extractall(LOCALAPPDATA / 'video2x' / 'waifu2x-converter-cpp')
-
-    def _install_waifu2x_ncnn_vulkan(self):
-        """ Install waifu2x-ncnn-vulkan
-        """
-        print('\nInstalling waifu2x-ncnn-vulkan')
-        import requests
-
-        # Get latest release of waifu2x-ncnn-vulkan via Github API
-        latest_release = requests.get('https://api.github.com/repos/nihui/waifu2x-ncnn-vulkan/releases/latest').json()
-
-        for a in latest_release['assets']:
-            if re.search(r'waifu2x-ncnn-vulkan-\d*\.zip', a['browser_download_url']):
-                waifu2x_ncnn_vulkan_zip = download(a['browser_download_url'], tempfile.gettempdir())
-                self.trash.append(waifu2x_ncnn_vulkan_zip)
-
-        # extract and rename
-        waifu2x_ncnn_vulkan_directory = LOCALAPPDATA / 'video2x' / 'waifu2x-ncnn-vulkan'
-        with zipfile.ZipFile(waifu2x_ncnn_vulkan_zip) as zipf:
-            zipf.extractall(LOCALAPPDATA / 'video2x')
-
-            # if directory already exists, remove it
-            if waifu2x_ncnn_vulkan_directory.exists():
-                shutil.rmtree(waifu2x_ncnn_vulkan_directory)
-
-            # rename the newly extracted directory
-            (LOCALAPPDATA / 'video2x' / zipf.namelist()[0]).rename(waifu2x_ncnn_vulkan_directory)
-
-    def _install_anime4kcpp(self):
-        """ Install Anime4KCPP
-        """
-        print('\nInstalling Anime4KCPP')
-
-        import pyunpack
-        import requests
-
-        # get latest release of Anime4KCPP via Github API
-        # at the time of writing this portion, Anime4KCPP doesn't yet have a stable release
-        # therefore releases/latest won't work
-        latest_release = requests.get('https://api.github.com/repos/TianZerL/Anime4KCPP/releases/latest').json()
-
-        for a in latest_release['assets']:
-            if re.search(r'Anime4KCPP_CLI-.*-Win64-msvc\.7z', a['browser_download_url']):
-                anime4kcpp_zip = download(a['browser_download_url'], tempfile.gettempdir())
-                self.trash.append(anime4kcpp_zip)
-
-        # extract and rename
-        # with py7zr.SevenZipFile(anime4kcpp_zip, mode='r') as archive:
-        (LOCALAPPDATA / 'video2x' / 'anime4kcpp').mkdir(parents=True, exist_ok=True)
-        pyunpack.Archive(anime4kcpp_zip).extractall(LOCALAPPDATA / 'video2x' / 'anime4kcpp')
-
-    def _install_srmd_ncnn_vulkan(self):
-        """ Install srmd-ncnn-vulkan
-        """
-        print('\nInstalling srmd-ncnn-vulkan')
-        import requests
-
-        # Get latest release of srmd-ncnn-vulkan via Github API
-        latest_release = requests.get('https://api.github.com/repos/nihui/srmd-ncnn-vulkan/releases/latest').json()
-
-        for a in latest_release['assets']:
-            if re.search(r'srmd-ncnn-vulkan-\d*\.zip', a['browser_download_url']):
-                srmd_ncnn_vulkan_zip = download(a['browser_download_url'], tempfile.gettempdir())
-                self.trash.append(srmd_ncnn_vulkan_zip)
-
-        # extract and rename
-        srmd_ncnn_vulkan_directory = LOCALAPPDATA / 'video2x' / 'srmd-ncnn-vulkan'
-        with zipfile.ZipFile(srmd_ncnn_vulkan_zip) as zipf:
-            zipf.extractall(LOCALAPPDATA / 'video2x')
-
-            # if directory already exists, remove it
-            if srmd_ncnn_vulkan_directory.exists():
-                shutil.rmtree(srmd_ncnn_vulkan_directory)
-
-            # rename the newly extracted directory
-            (LOCALAPPDATA / 'video2x' / zipf.namelist()[0]).rename(srmd_ncnn_vulkan_directory)
-
-    def _generate_config(self):
-        """ Generate video2x config
-        """
-        import yaml
-
-        # open current video2x configuration file as template
-        with open(VIDEO2X_CONFIG, 'r') as template:
-            template_dict = yaml.load(template, Loader=yaml.FullLoader)
-            template.close()
-
-        # configure only the specified drivers
-        if self.driver == 'all':
-            template_dict['waifu2x_caffe']['path'] = str(LOCALAPPDATA / 'video2x' / 'waifu2x-caffe' / 'waifu2x-caffe-cui')
-            template_dict['waifu2x_converter_cpp']['path'] = str(LOCALAPPDATA / 'video2x' / 'waifu2x-converter-cpp' / 'waifu2x-converter-cpp')
-            template_dict['waifu2x_ncnn_vulkan']['path'] = str(LOCALAPPDATA / 'video2x' / 'waifu2x-ncnn-vulkan' / 'waifu2x-ncnn-vulkan')
-            template_dict['srmd_ncnn_vulkan']['path'] = str(LOCALAPPDATA / 'video2x' / 'srmd-ncnn-vulkan' / 'srmd-ncnn-vulkan')
-            template_dict['anime4kcpp']['path'] = str(LOCALAPPDATA / 'video2x' / 'anime4kcpp' / 'CLI' / 'Anime4KCPP_CLI' / 'Anime4KCPP_CLI')
-        elif self.driver == 'waifu2x_caffe':
-            template_dict['waifu2x_caffe']['path'] = str(LOCALAPPDATA / 'video2x' / 'waifu2x-caffe' / 'waifu2x-caffe-cui')
-        elif self.driver == 'waifu2x_converter_cpp':
-            template_dict['waifu2x_converter_cpp']['path'] = str(LOCALAPPDATA / 'video2x' / 'waifu2x-converter-cpp' / 'waifu2x-converter-cpp')
-        elif self.driver == 'waifu2x_ncnn_vulkan':
-            template_dict['waifu2x_ncnn_vulkan']['path'] = str(LOCALAPPDATA / 'video2x' / 'waifu2x-ncnn-vulkan' / 'waifu2x-ncnn-vulkan')
-        elif self.driver == 'srmd_ncnn_vulkan':
-            template_dict['srmd_ncnn_vulkan']['path'] = str(LOCALAPPDATA / 'video2x' / 'srmd-ncnn-vulkan' / 'srmd-ncnn-vulkan')
-        elif self.driver == 'anime4kcpp':
-            template_dict['anime4kcpp']['path'] = str(LOCALAPPDATA / 'video2x' / 'anime4kcpp' / 'CLI' / 'Anime4KCPP_CLI' / 'Anime4KCPP_CLI')
-
-        template_dict['ffmpeg']['ffmpeg_path'] = str(LOCALAPPDATA / 'video2x' / 'ffmpeg-latest-win64-static' / 'bin')
-        template_dict['video2x']['video2x_cache_directory'] = None
-        template_dict['video2x']['preserve_frames'] = False
-
-        # write configuration into file
-        with open(VIDEO2X_CONFIG, 'w') as config:
-            yaml.dump(template_dict, config)
-
-
-def download(url, save_path, chunk_size=4096):
-    """ Download file to local with requests library
-    """
-    from tqdm import tqdm
-    import requests
-
-    save_path = pathlib.Path(save_path)
-
-    # create target folder if it doesn't exist
-    save_path.mkdir(parents=True, exist_ok=True)
-
-    # create requests stream for steaming file
-    stream = requests.get(url, stream=True, allow_redirects=True)
-
-    # get file name
-    file_name = None
-    if 'content-disposition' in stream.headers:
-        disposition = stream.headers['content-disposition']
-        with contextlib.suppress(IndexError):
-            file_name = re.findall("filename=(.+)", disposition)[0].strip('"')
-
-    if file_name is None:
-        # output_file = f'{save_path}\\{stream.url.split("/")[-1]}'
-        output_file = save_path / stream.url.split('/')[-1]
-    else:
-        output_file = save_path / file_name
-
-    # decode url encoding
-    output_file = pathlib.Path(urllib.parse.unquote(str(output_file)))
-
-    # get total size for progress bar if provided in headers
-    total_size = 0
-    if 'content-length' in stream.headers:
-        total_size = int(stream.headers['content-length'])
-
-    # print download information summary
-    print(f'Downloading: {url}')
-    print(f'Total size: {total_size}')
-    print(f'Chunk size: {chunk_size}')
-    print(f'Saving to: {output_file}')
-
-    # Write content into file
-    with open(output_file, 'wb') as output:
-        with tqdm(total=total_size, ascii=True) as progress_bar:
-            for chunk in stream.iter_content(chunk_size=chunk_size):
-                if chunk:
-                    output.write(chunk)
-                    progress_bar.update(len(chunk))
-
-    # return the full path of saved file
-    return output_file
-
-
-def pip_install(file):
-    """ Install python package via python pip module
-
-    pip.main() is not available after pip 9.0.1, thus
-    pip module is not used in this case.
-    """
-    return subprocess.run([sys.executable, '-m', 'pip', 'install', '-U', '-r', file]).returncode
-
-
-if __name__ == '__main__':
-    try:
-        # set default exit code
-        EXIT_CODE = 0
-
-        # get start time
-        start_time = time.time()
-
-        # check platform
-        if sys.platform != 'win32':
-            print('This script is currently only compatible with Windows')
-            EXIT_CODE = 1
-            sys.exit(1)
-
-        # parse command line arguments
-        args = parse_arguments()
-        print('Video2X Setup Script')
-        print(f'Version: {VERSION}')
-
-        # do not install pip modules if script
-        # is packaged in exe format
-        download_python_modules = True
-        if sys.argv[0].endswith('.exe'):
-            print('\nScript is packaged as exe, skipping pip module download')
-            download_python_modules = False
-
-        setup = Video2xSetup(args.driver, download_python_modules)
-        setup.run()
-        print('\nScript finished successfully')
-
-    except SystemExit:
-        pass
-
-    # if PermissionError is raised
-    # user needs to run this with higher privilege
-    except PermissionError:
-        traceback.print_exc()
-        print('You might have insufficient privilege for this script to run')
-        print('Try running this script with Administrator privileges')
-        EXIT_CODE = 1
-
-    # for any exception in the script
-    except Exception:
-        traceback.print_exc()
-        print('An error has occurred')
-        print('Video2X Automatic Setup has failed')
-
-        # in case of a failure, try cleaning up temp files
-        try:
-            setup._cleanup()
-        except Exception:
-            traceback.print_exc()
-            print('An error occurred while trying to cleanup files')
-
-        EXIT_CODE = 1
-
-    # regardless if script finishes successfully or not
-    # print script execution summary
-    finally:
-        print('Script finished')
-        print(f'Time taken: {timedelta(seconds=round(time.time() - start_time))}')
-        input('Press [ENTER] to exit script')
-        sys.exit(EXIT_CODE)

src/wrappers/anime4kcpp.py

@@ -1,116 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Name: Waifu2x Caffe Driver
-Author: K4YT3X
-Date Created: May 3, 2020
-Last Modified: May 11, 2020
-
-Description: This class is a high-level wrapper
-for waifu2x-caffe.
-"""
-
-# built-in imports
-import argparse
-import os
-import pathlib
-import platform
-import shlex
-import subprocess
-import threading
-
-# third-party imports
-from avalon_framework import Avalon
-
-
-class WrapperMain:
-    """ Anime4K CPP wrapper
-    """
-
-    def __init__(self, driver_settings):
-        self.driver_settings = driver_settings
-        self.print_lock = threading.Lock()
-
-    @staticmethod
-    def zero_to_one_float(value):
-        value = float(value)
-        if value < 0.0 or value > 1.0:
-            raise argparse.ArgumentTypeError(f'{value} is not between 0.0 and 1.0')
-        return value
-
-    @staticmethod
-    def parse_arguments(arguments):
-        parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter, add_help=False)
-        parser.error = lambda message: (_ for _ in ()).throw(AttributeError(message))
-        parser.add_argument('--help', action='help', help='show this help message and exit')
-        parser.add_argument('-i', '--input', type=str, help=argparse.SUPPRESS)  # help='File for loading')
-        parser.add_argument('-o', '--output', type=str, help=argparse.SUPPRESS)  # help='File for outputting')
-        parser.add_argument('-p', '--passes', type=int, help='Passes for processing')
-        parser.add_argument('-n', '--pushColorCount', type=int, help='Limit the number of color pushes')
-        parser.add_argument('-c', '--strengthColor', type=WrapperMain.zero_to_one_float, help='Strength for pushing color,range 0 to 1,higher for thinner')
-        parser.add_argument('-g', '--strengthGradient', type=WrapperMain.zero_to_one_float, help='Strength for pushing gradient,range 0 to 1,higher for sharper')
-        parser.add_argument('-z', '--zoomFactor', type=float, help='zoom factor for resizing')
-        parser.add_argument('-t', '--threads', type=int, help='Threads count for video processing')
-        parser.add_argument('-f', '--fastMode', action='store_true', help='Faster but maybe low quality')
-        parser.add_argument('-v', '--videoMode', action='store_true', help='Video process')
-        parser.add_argument('-s', '--preview', action='store_true', help='Preview image')
-        parser.add_argument('-b', '--preprocessing', action='store_true', help='Enable pre processing')
-        parser.add_argument('-a', '--postprocessing', action='store_true', help='Enable post processing')
-        parser.add_argument('-r', '--preFilters', type=int, help='Enhancement filter, only working when preProcessing is true,there are 5 options by binary:Median blur=0000001, Mean blur=0000010, CAS Sharpening=0000100, Gaussian blur weak=0001000, Gaussian blur=0010000, Bilateral filter=0100000, Bilateral filter faster=1000000, you can freely combine them, eg: Gaussian blur weak + Bilateral filter = 0001000 | 0100000 = 0101000 = 40(D)')
-        parser.add_argument('-e', '--postFilters', type=int, help='Enhancement filter, only working when postProcessing is true,there are 5 options by binary:Median blur=0000001, Mean blur=0000010, CAS Sharpening=0000100, Gaussian blur weak=0001000, Gaussian blur=0010000, Bilateral filter=0100000, Bilateral filter faster=1000000, you can freely combine them, eg: Gaussian blur weak + Bilateral filter = 0001000 | 0100000 = 0101000 = 40(D), so you can put 40 to enable Gaussian blur weak and Bilateral filter, which also is what I recommend for image that < 1080P, 48 for image that >= 1080P, and for performance I recommend to use 72 for video that < 1080P, 80 for video that >=1080P')
-        parser.add_argument('-q', '--GPUMode', action='store_true', help='Enable GPU acceleration')
-        parser.add_argument('-l', '--listGPUs', action='store_true', help='list GPUs')
-        parser.add_argument('-h', '--platformID', type=int, help='Specify the platform ID')
-        parser.add_argument('-d', '--deviceID', type=int, help='Specify the device ID')
-        parser.add_argument('-C', '--codec', type=str, help='Specify the codec for encoding from mp4v(recommended in Windows), dxva(for Windows), avc1(H264, recommended in Linux), vp09(very slow), hevc(not support in Windowds), av01(not support in Windowds) (string [=mp4v])')
-        return parser.parse_args(arguments)
-
-    def load_configurations(self, upscaler):
-        self.driver_settings['zoomFactor'] = upscaler.scale_ratio
-        self.driver_settings['threads'] = upscaler.processes
-
-    def upscale(self, input_file, output_file):
-        """This is the core function for WAIFU2X class
-
-        Arguments:
-            input_file {string} -- source directory path
-            output_file {string} -- output directory path
-            width {int} -- output video width
-            height {int} -- output video height
-        """
-
-        # overwrite config file settings
-        self.driver_settings['input'] = input_file
-        self.driver_settings['output'] = output_file
-
-        # Anime4KCPP will look for Anime4KCPPKernel.cl under the current working directory
-        # change the CWD to its containing directory so it will find it
-        if platform.system() == 'Windows':
-            os.chdir(pathlib.Path(self.driver_settings['path']).parent)
-
-        # list to be executed
-        # initialize the list with waifu2x binary path as the first element
-        execute = [self.driver_settings.pop('path')]
-
-        for key in self.driver_settings.keys():
-
-            value = self.driver_settings[key]
-
-            # null or None means that leave this option out (keep default)
-            if value is None or value is False:
-                continue
-            else:
-                if len(key) == 1:
-                    execute.append(f'-{key}')
-                else:
-                    execute.append(f'--{key}')
-
-                # true means key is an option
-                if value is not True:
-                    execute.append(str(value))
-
-        # return the Popen object of the new process created
-        self.print_lock.acquire()
-        Avalon.debug_info(f'[upscaler] Subprocess {os.getpid()} executing: {shlex.join(execute)}')
-        self.print_lock.release()
-        return subprocess.Popen(execute)

src/wrappers/ffmpeg.py

@@ -1,290 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Name: Video2X FFmpeg Controller
-Author: K4YT3X
-Date Created: Feb 24, 2018
-Last Modified: May 9, 2020
-
-Description: This class handles all FFmpeg related operations.
-"""
-
-# built-in imports
-import json
-import pathlib
-import subprocess
-
-# third-party imports
-from avalon_framework import Avalon
-
-
-class Ffmpeg:
-    """This class communicates with FFmpeg
-
-    This class deals with FFmpeg. It handles extracting
-    frames, stripping audio, converting images into videos
-    and inserting audio tracks to videos.
-    """
-
-    def __init__(self, ffmpeg_settings, image_format):
-        self.ffmpeg_settings = ffmpeg_settings
-
-        self.ffmpeg_path = pathlib.Path(self.ffmpeg_settings['ffmpeg_path'])
-        self.ffmpeg_binary = self.ffmpeg_path / 'ffmpeg'
-        self.ffmpeg_probe_binary = self.ffmpeg_path / 'ffprobe'
-
-        # video metadata
-        self.image_format = image_format
-        self.intermediate_file_name = pathlib.Path(self.ffmpeg_settings['intermediate_file_name'])
-        self.pixel_format = None
-
-    def get_pixel_formats(self):
-        """ Get a dictionary of supported pixel formats
-
-        List all supported pixel formats and their
-        corresponding bit depth.
-
-        Returns:
-            dictionary -- JSON dict of all pixel formats to bit depth
-        """
-        execute = [
-            self.ffmpeg_probe_binary,
-            '-v',
-            'quiet',
-            '-pix_fmts'
-        ]
-
-        # turn elements into str
-        execute = [str(e) for e in execute]
-
-        Avalon.debug_info(f'Executing: {" ".join(execute)}')
-
-        # initialize dictionary to store pixel formats
-        pixel_formats = {}
-
-        # record all pixel formats into dictionary
-        for line in subprocess.run(execute, check=True, stdout=subprocess.PIPE).stdout.decode().split('\n'):
-            try:
-                pixel_formats[' '.join(line.split()).split()[1]] = int(' '.join(line.split()).split()[3])
-            except (IndexError, ValueError):
-                pass
-
-        # print pixel formats for debugging
-        Avalon.debug_info(str(pixel_formats))
-
-        return pixel_formats
-
-    def get_video_info(self, input_video):
-        """ Gets input video information
-
-        This method reads input video information
-        using ffprobe in dictionary
-
-        Arguments:
-            input_video {string} -- input video file path
-
-        Returns:
-            dictionary -- JSON text of input video information
-        """
-
-        # this execution command needs to be hard-coded
-        # since video2x only strictly recignizes this one format
-        execute = [
-            self.ffmpeg_probe_binary,
-            '-v',
-            'quiet',
-            '-print_format',
-            'json',
-            '-show_format',
-            '-show_streams',
-            '-i',
-            input_video
-        ]
-
-        # turn elements into str
-        execute = [str(e) for e in execute]
-
-        Avalon.debug_info(f'Executing: {" ".join(execute)}')
-        json_str = subprocess.run(execute, check=True, stdout=subprocess.PIPE).stdout
-        return json.loads(json_str.decode('utf-8'))
-
-    def extract_frames(self, input_video, extracted_frames):
-        """Extract every frame from original videos
-
-        This method extracts every frame from input video using FFmpeg
-
-        Arguments:
-            input_video {string} -- input video path
-            extracted_frames {string} -- video output directory
-        """
-        execute = [
-            self.ffmpeg_binary
-        ]
-
-        execute.extend(self._read_configuration(phase='video_to_frames'))
-
-        execute.extend([
-            '-i',
-            input_video
-        ])
-
-        execute.extend(self._read_configuration(phase='video_to_frames', section='output_options'))
-
-        execute.extend([
-            extracted_frames / f'extracted_%0d.{self.image_format}'
-        ])
-
-        return(self._execute(execute))
-
-    def assemble_video(self, framerate, resolution, upscaled_frames):
-        """Converts images into videos
-
-        This method converts a set of images into a video
-
-        Arguments:
-            framerate {float} -- target video framerate
-            resolution {string} -- target video resolution
-            upscaled_frames {string} -- source images directory
-        """
-        execute = [
-            self.ffmpeg_binary,
-            '-r',
-            str(framerate),
-            '-s',
-            resolution
-        ]
-
-        # read other options
-        execute.extend(self._read_configuration(phase='frames_to_video'))
-
-        # read FFmpeg input options
-        execute.extend(self._read_configuration(phase='frames_to_video', section='input_options'))
-
-        # WORKAROUND FOR WAIFU2X-NCNN-VULKAN
-        # Dev: SAT3LL
-        # rename all .png.png suffixes to .png
-        import re
-        regex = re.compile(r'\.png\.png$', re.IGNORECASE)
-        for frame_name in upscaled_frames.iterdir():
-            (upscaled_frames / frame_name).rename(upscaled_frames / regex.sub('.png', str(frame_name)))
-        # END WORKAROUND
-
-        # append input frames path into command
-        execute.extend([
-            '-i',
-            upscaled_frames / f'extracted_%d.{self.image_format}'
-        ])
-
-        # read FFmpeg output options
-        execute.extend(self._read_configuration(phase='frames_to_video', section='output_options'))
-
-        # specify output file location
-        execute.extend([
-            upscaled_frames / self.intermediate_file_name
-        ])
-
-        return(self._execute(execute))
-
-    def migrate_streams(self, input_video, output_video, upscaled_frames):
-        """ Migrates audio tracks and subtitles from input video to output video
-
-        Arguments:
-            input_video {string} -- input video file path
-            output_video {string} -- output video file path
-            upscaled_frames {string} -- directory containing upscaled frames
-        """
-        execute = [
-            self.ffmpeg_binary
-        ]
-
-        execute.extend(self._read_configuration(phase='migrating_tracks'))
-
-        execute.extend([
-            '-i',
-            upscaled_frames / self.intermediate_file_name,
-            '-i',
-            input_video
-        ])
-
-        execute.extend(self._read_configuration(phase='migrating_tracks', section='output_options'))
-
-        execute.extend([
-            output_video
-        ])
-
-        return(self._execute(execute))
-
-    def _read_configuration(self, phase, section=None):
-        """ read configuration from JSON
-
-        Read the configurations (arguments) from the JSON
-        configuration file and append them to the end of the
-        FFmpeg command.
-
-        Arguments:
-            execute {list} -- list of arguments to be executed
-            phase {str} -- phase of operation
-        """
-
-        configuration = []
-
-        # if section is specified, read configurations or keys
-        # from only that section
-        if section:
-            source = self.ffmpeg_settings[phase][section].keys()
-
-            # if pixel format is not specified, use the source pixel format
-            try:
-                if self.ffmpeg_settings[phase][section].get('-pix_fmt') is None:
-                    self.ffmpeg_settings[phase][section]['-pix_fmt'] = self.pixel_format
-            except KeyError:
-                pass
-        else:
-            source = self.ffmpeg_settings[phase].keys()
-
-        for key in source:
-
-            if section:
-                value = self.ffmpeg_settings[phase][section][key]
-            else:
-                value = self.ffmpeg_settings[phase][key]
-
-            # null or None means that leave this option out (keep default)
-            if value is None or value is False or isinstance(value, dict):
-                continue
-
-            # if the value is a list, append the same argument and all values
-            elif isinstance(value, list):
-
-                for subvalue in value:
-                    configuration.append(key)
-                    if value is not True:
-                        configuration.append(str(subvalue))
-
-            # otherwise the value is typical
-            else:
-                configuration.append(key)
-
-                # true means key is an option
-                if value is True:
-                    continue
-
-                configuration.append(str(value))
-
-        return configuration
-
-    def _execute(self, execute):
-        """ execute command
-
-        Arguments:
-            execute {list} -- list of arguments to be executed
-
-        Returns:
-            int -- execution return code
-        """
-        # turn all list elements into string to avoid errors
-        execute = [str(e) for e in execute]
-
-        Avalon.debug_info(f'Executing: {execute}')
-
-        return subprocess.Popen(execute)

src/wrappers/srmd_ncnn_vulkan.py

@@ -1,103 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Name: SRMD NCNN Vulkan Driver
-Creator: K4YT3X
-Date Created: April 26, 2020
-Last Modified: May 11, 2020
-
-Description: This class is a high-level wrapper
-for srmd_ncnn_vulkan.
-"""
-
-# built-in imports
-import argparse
-import os
-import pathlib
-import platform
-import shlex
-import subprocess
-import threading
-
-# third-party imports
-from avalon_framework import Avalon
-
-
-class WrapperMain:
-    """This class communicates with SRMD NCNN Vulkan engine
-
-    An object will be created for this class, containing information
-    about the binary address and the processing method. When being called
-    by the main program, other detailed information will be passed to
-    the upscale function.
-    """
-
-    def __init__(self, driver_settings):
-        self.driver_settings = driver_settings
-        self.print_lock = threading.Lock()
-
-    @staticmethod
-    def parse_arguments(arguments):
-        parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter, add_help=False)
-        parser.error = lambda message: (_ for _ in ()).throw(AttributeError(message))
-        parser.add_argument('--help', action='help', help='show this help message and exit')
-        parser.add_argument('-v', action='store_true', help='verbose output')
-        parser.add_argument('-i', type=str, help=argparse.SUPPRESS)  # help='input image path (jpg/png) or directory')
-        parser.add_argument('-o', type=str, help=argparse.SUPPRESS)  # help='output image path (png) or directory')
-        parser.add_argument('-n', type=int, choices=range(-1, 11), help='denoise level')
-        parser.add_argument('-s', type=int, choices=range(2, 5), help='upscale ratio')
-        parser.add_argument('-t', type=int, help='tile size (>=32)')
-        parser.add_argument('-m', type=str, help='srmd model path')
-        parser.add_argument('-g', type=int, help='gpu device to use')
-        parser.add_argument('-j', type=str, help='thread count for load/proc/save')
-        parser.add_argument('-x', action='store_true', help='enable tta mode')
-        return parser.parse_args(arguments)
-
-    def load_configurations(self, upscaler):
-        self.driver_settings['s'] = int(upscaler.scale_ratio)
-        self.driver_settings['j'] = '{}:{}:{}'.format(upscaler.processes, upscaler.processes, upscaler.processes)
-
-    def upscale(self, input_directory, output_directory):
-        """This is the core function for SRMD ncnn Vulkan class
-
-        Arguments:
-            input_directory {string} -- source directory path
-            output_directory {string} -- output directory path
-            ratio {int} -- output video ratio
-        """
-
-        # overwrite config file settings
-        self.driver_settings['i'] = input_directory
-        self.driver_settings['o'] = output_directory
-
-        # by default, srmd-ncnn-vulkan will look for the models under the current working directory
-        # change the working directory to its containing folder if model directory not specified
-        if self.driver_settings['m'] is None and platform.system() == 'Windows':
-            os.chdir(pathlib.Path(self.driver_settings['path']).parent)
-
-        # list to be executed
-        # initialize the list with the binary path as the first element
-        execute = [self.driver_settings.pop('path')]
-
-        for key in self.driver_settings.keys():
-
-            value = self.driver_settings[key]
-
-            # null or None means that leave this option out (keep default)
-            if value is None or value is False:
-                continue
-            else:
-                if len(key) == 1:
-                    execute.append(f'-{key}')
-                else:
-                    execute.append(f'--{key}')
-
-                # true means key is an option
-                if value is not True:
-                    execute.append(str(value))
-
-        # return the Popen object of the new process created
-        self.print_lock.acquire()
-        Avalon.debug_info(f'[upscaler] Subprocess {os.getpid()} executing: {shlex.join(execute)}')
-        self.print_lock.release()
-        return subprocess.Popen(execute)

src/wrappers/waifu2x_caffe.py

@@ -1,105 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Name: Waifu2x Caffe Driver
-Author: K4YT3X
-Date Created: Feb 24, 2018
-Last Modified: May 11, 2020
-
-Description: This class is a high-level wrapper
-for waifu2x-caffe.
-"""
-
-# built-in imports
-import argparse
-import os
-import shlex
-import subprocess
-import threading
-
-# third-party imports
-from avalon_framework import Avalon
-
-
-class WrapperMain:
-    """This class communicates with waifu2x cui engine
-
-    An object will be created for this class, containing information
-    about the binary address and the processing method. When being called
-    by the main program, other detailed information will be passed to
-    the upscale function.
-    """
-
-    def __init__(self, driver_settings):
-        self.driver_settings = driver_settings
-        self.print_lock = threading.Lock()
-
-    @staticmethod
-    def parse_arguments(arguments):
-        parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter, add_help=False)
-        parser.error = lambda message: (_ for _ in ()).throw(AttributeError(message))
-        parser.add_argument('--help', action='help', help='show this help message and exit')
-        parser.add_argument('-t', '--tta', type=int, choices=range(2), help='8x slower and slightly high quality')
-        parser.add_argument('--gpu', type=int, help='gpu device no')
-        parser.add_argument('-b', '--batch_size', type=int, help='input batch size')
-        parser.add_argument('--crop_h', type=int, help='input image split size(height)')
-        parser.add_argument('--crop_w', type=int, help='input image split size(width)')
-        parser.add_argument('-c', '--crop_size', type=int, help='input image split size')
-        parser.add_argument('-d', '--output_depth', type=int, help='output image chaneel depth bit')
-        parser.add_argument('-q', '--output_quality', type=int, help='output image quality')
-        parser.add_argument('-p', '--process', choices=['cpu', 'gpu', 'cudnn'], help='process mode')
-        parser.add_argument('--model_dir', type=str, help='path to custom model directory (don\'t append last / )')
-        parser.add_argument('-h', '--scale_height', type=int, help='custom scale height')
-        parser.add_argument('-w', '--scale_width', type=int, help='custom scale width')
-        parser.add_argument('-s', '--scale_ratio', type=float, help='custom scale ratio')
-        parser.add_argument('-n', '--noise_level', type=int, choices=range(4), help='noise reduction level')
-        parser.add_argument('-m', '--mode', choices=['noise', 'scale', 'noise_scale', 'auto_scale'], help='image processing mode')
-        parser.add_argument('-e', '--output_extention', type=str, help='extention to output image file when output_path is (auto) or input_path is folder')
-        parser.add_argument('-l', '--input_extention_list', type=str, help='extention to input image file when input_path is folder')
-        parser.add_argument('-o', '--output', type=str, help=argparse.SUPPRESS)  # help='path to output image file (when input_path is folder, output_path must be folder)')
-        parser.add_argument('-i', '--input_file', type=str, help=argparse.SUPPRESS)  # help='(required) path to input image file')
-        return parser.parse_args(arguments)
-
-    def load_configurations(self, upscaler):
-        # use scale width and scale height if specified
-        self.driver_settings['scale_ratio'] = upscaler.scale_ratio
-        self.driver_settings['output_extention'] = upscaler.image_format
-
-        # bit_depth will be 12 at this point
-        # it will up updated later
-        self.driver_settings['output_depth'] = 12
-
-    def upscale(self, input_directory, output_directory):
-        """ start upscaling process
-        """
-
-        # overwrite config file settings
-        self.driver_settings['input_path'] = input_directory
-        self.driver_settings['output_path'] = output_directory
-
-        # list to be executed
-        # initialize the list with waifu2x binary path as the first element
-        execute = [self.driver_settings.pop('path')]
-
-        for key in self.driver_settings.keys():
-
-            value = self.driver_settings[key]
-
-            # null or None means that leave this option out (keep default)
-            if value is None or value is False:
-                continue
-            else:
-                if len(key) == 1:
-                    execute.append(f'-{key}')
-                else:
-                    execute.append(f'--{key}')
-
-                # true means key is an option
-                if value is not True:
-                    execute.append(str(value))
-
-        # return the Popen object of the new process created
-        self.print_lock.acquire()
-        Avalon.debug_info(f'[upscaler] Subprocess {os.getpid()} executing: {shlex.join(execute)}')
-        self.print_lock.release()
-        return subprocess.Popen(execute)

src/wrappers/waifu2x_converter_cpp.py

@@ -1,119 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Name: Waifu2x Converter CPP Driver
-Author: K4YT3X
-Date Created: February 8, 2019
-Last Modified: May 11, 2020
-
-Description: This class is a high-level wrapper
-for waifu2x-converter-cpp.
-"""
-
-# built-in imports
-import argparse
-import os
-import pathlib
-import shlex
-import subprocess
-import threading
-
-# third-party imports
-from avalon_framework import Avalon
-
-
-class WrapperMain:
-    """This class communicates with waifu2x cui engine
-
-    An object will be created for this class, containing information
-    about the binary address and the processing method. When being called
-    by the main program, other detailed information will be passed to
-    the upscale function.
-    """
-
-    def __init__(self, driver_settings):
-        self.driver_settings = driver_settings
-        self.print_lock = threading.Lock()
-
-    @staticmethod
-    def parse_arguments(arguments):
-        parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter, add_help=False)
-        parser.error = lambda message: (_ for _ in ()).throw(AttributeError(message))
-        parser.add_argument('--help', action='help', help='show this help message and exit')
-        parser.add_argument('--list-supported-formats', action='store_true', help='dump currently supported format list')
-        parser.add_argument('--list-opencv-formats', action='store_true', help='(deprecated. Use --list-supported-formats) dump opencv supported format list')
-        parser.add_argument('-l', '--list-processor', action='store_true', help='dump processor list')
-        parser.add_argument('-f', '--output-format', choices=['png', 'jpg'], help='The format used when running in recursive/folder mode\nSee --list-supported-formats for a list of supported formats/extensions.')
-        parser.add_argument('-c', '--png-compression', type=int, choices=range(10), help='Set PNG compression level (0-9), 9 = Max compression (slowest & smallest)')
-        parser.add_argument('-q', '--image-quality', type=int, choices=range(-1, 102), help='JPEG & WebP Compression quality (0-101, 0 being smallest size and lowest quality), use 101 for lossless WebP')
-        parser.add_argument('--block-size', type=int, help='block size')
-        parser.add_argument('--disable-gpu', action='store_true', help='disable GPU')
-        parser.add_argument('--force-OpenCL', action='store_true', help='force to use OpenCL on Intel Platform')
-        parser.add_argument('-p', '--processor', type=int, help='set target processor')
-        parser.add_argument('-j', '--jobs', type=int, help='number of threads launching at the same time')
-        parser.add_argument('--model-dir', type=str, help='path to custom model directory (don\'t append last / )')
-        parser.add_argument('--scale-ratio', type=float, help='custom scale ratio')
-        parser.add_argument('--noise-level', type=int, choices=range(4), help='noise reduction level')
-        parser.add_argument('-m', '--mode', choices=['noise', 'scale', 'noise-scale'], help='image processing mode')
-        parser.add_argument('-v', '--log-level', type=int, choices=range(5), help='Set log level')
-        parser.add_argument('-s', '--silent', action='store_true', help='Enable silent mode. (same as --log-level 1)')
-        parser.add_argument('-t', '--tta', type=int, choices=range(2), help='Enable Test-Time Augmentation mode.')
-        parser.add_argument('-g', '--generate-subdir', type=int, choices=range(2), help='Generate sub folder when recursive directory is enabled.')
-        parser.add_argument('-a', '--auto-naming', type=int, choices=range(2), help='Add postfix to output name when output path is not specified.\nSet 0 to disable this.')
-        parser.add_argument('-r', '--recursive-directory', type=int, choices=range(2), help='Search recursively through directories to find more images to process.')
-        parser.add_argument('-o', '--output', type=str, help=argparse.SUPPRESS)  # help='path to output image file or directory  (you should use the full path)')
-        parser.add_argument('-i', '--input', type=str, help=argparse.SUPPRESS)  # help='(required)  path to input image file or directory (you should use the full path)')
-        parser.add_argument('--version', action='store_true', help='Displays version information and exits.')
-        return parser.parse_args(arguments)
-
-    def load_configurations(self, upscaler):
-        self.driver_settings['scale-ratio'] = upscaler.scale_ratio
-        self.driver_settings['jobs'] = upscaler.processes
-        self.driver_settings['output-format'] = upscaler.image_format.lower()
-
-    def upscale(self, input_directory, output_directory):
-        """ Waifu2x Converter Driver Upscaler
-        This method executes the upscaling of extracted frames.
-
-        Arguments:
-            input_directory {string} -- source directory path
-            output_directory {string} -- output directory path
-            scale_ratio {int} -- frames' scale ratio
-            threads {int} -- number of threads
-        """
-
-        # overwrite config file settings
-        self.driver_settings['input'] = input_directory
-        self.driver_settings['output'] = output_directory
-
-        # models_rgb must be specified manually for waifu2x-converter-cpp
-        # if it's not specified in the arguments, create automatically
-        if self.driver_settings['model-dir'] is None:
-            self.driver_settings['model-dir'] = pathlib.Path(self.driver_settings['path']).parent / 'models_rgb'
-
-        # list to be executed
-        # initialize the list with waifu2x binary path as the first element
-        execute = [self.driver_settings.pop('path')]
-
-        for key in self.driver_settings.keys():
-
-            value = self.driver_settings[key]
-
-            # null or None means that leave this option out (keep default)
-            if value is None or value is False:
-                continue
-            else:
-                if len(key) == 1:
-                    execute.append(f'-{key}')
-                else:
-                    execute.append(f'--{key}')
-
-                # true means key is an option
-                if value is not True:
-                    execute.append(str(value))
-
-        # return the Popen object of the new process created
-        self.print_lock.acquire()
-        Avalon.debug_info(f'[upscaler] Subprocess {os.getpid()} executing: {shlex.join(execute)}')
-        self.print_lock.release()
-        return subprocess.Popen(execute)

src/wrappers/waifu2x_ncnn_vulkan.py

@@ -1,106 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-Name: Waifu2x NCNN Vulkan Driver
-Creator: SAT3LL
-Date Created: June 26, 2019
-Last Modified: May 11, 2020
-
-Editor: K4YT3X
-Last Modified: February 22, 2020
-
-Description: This class is a high-level wrapper
-for waifu2x_ncnn_vulkan.
-"""
-
-# built-in imports
-import argparse
-import os
-import pathlib
-import platform
-import shlex
-import subprocess
-import threading
-
-# third-party imports
-from avalon_framework import Avalon
-
-
-class WrapperMain:
-    """This class communicates with waifu2x ncnn vulkan engine
-
-    An object will be created for this class, containing information
-    about the binary address and the processing method. When being called
-    by the main program, other detailed information will be passed to
-    the upscale function.
-    """
-
-    def __init__(self, driver_settings):
-        self.driver_settings = driver_settings
-        self.print_lock = threading.Lock()
-
-    @staticmethod
-    def parse_arguments(arguments):
-        parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter, add_help=False)
-        parser.error = lambda message: (_ for _ in ()).throw(AttributeError(message))
-        parser.add_argument('--help', action='help', help='show this help message and exit')
-        parser.add_argument('-v', action='store_true', help='verbose output')
-        parser.add_argument('-i', type=str, help=argparse.SUPPRESS)  # help='input image path (jpg/png) or directory')
-        parser.add_argument('-o', type=str, help=argparse.SUPPRESS)  # help='output image path (png) or directory')
-        parser.add_argument('-n', type=int, choices=range(-1, 4), help='denoise level')
-        parser.add_argument('-s', type=int, choices=range(1, 3), help='upscale ratio')
-        parser.add_argument('-t', type=int, help='tile size (>=32)')
-        parser.add_argument('-m', type=str, help='waifu2x model path')
-        parser.add_argument('-g', type=int, help='gpu device to use')
-        parser.add_argument('-j', type=str, help='thread count for load/proc/save')
-        parser.add_argument('-x', action='store_true', help='enable tta mode')
-        return parser.parse_args(arguments)
-
-    def load_configurations(self, upscaler):
-        self.driver_settings['s'] = int(upscaler.scale_ratio)
-        self.driver_settings['j'] = '{}:{}:{}'.format(upscaler.processes, upscaler.processes, upscaler.processes)
-
-    def upscale(self, input_directory, output_directory):
-        """This is the core function for WAIFU2X class
-
-        Arguments:
-            input_directory {string} -- source directory path
-            output_directory {string} -- output directory path
-            ratio {int} -- output video ratio
-        """
-
-        # overwrite config file settings
-        self.driver_settings['i'] = input_directory
-        self.driver_settings['o'] = output_directory
-
-        # by default, waifu2x-ncnn-vulkan will look for the models under the current working directory
-        # change the working directory to its containing folder if model directory not specified
-        if self.driver_settings['m'] is None and platform.system() == 'Windows':
-            os.chdir(pathlib.Path(self.driver_settings['path']).parent)
-
-        # list to be executed
-        # initialize the list with waifu2x binary path as the first element
-        execute = [self.driver_settings.pop('path')]
-
-        for key in self.driver_settings.keys():
-
-            value = self.driver_settings[key]
-
-            # null or None means that leave this option out (keep default)
-            if value is None or value is False:
-                continue
-            else:
-                if len(key) == 1:
-                    execute.append(f'-{key}')
-                else:
-                    execute.append(f'--{key}')
-
-                # true means key is an option
-                if value is not True:
-                    execute.append(str(value))
-
-        # return the Popen object of the new process created
-        self.print_lock.acquire()
-        Avalon.debug_info(f'[upscaler] Subprocess {os.getpid()} executing: {shlex.join(execute)}')
-        self.print_lock.release()
-        return subprocess.Popen(execute)