Ma/video-subtitle-remover
1
0
Fork 0
mirror of https://github.com/YaoFANGUK/video-subtitle-remover.git synced 2026-05-07 10:07:30 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

使用PySide6-Fluent-Widgets重构整套UI

Browse Source 
添加任务列表组件并优化视频加载逻辑
支持可视化显示字幕区域
整理所有模型, 分别为STTN智能擦除, STTN字幕检测, LAMA, ProPainter, OpenCV
提高处理性能
新增CPU运行模式并优化多语言支持
修复Propainter模式部分视频报错

本次提交新增了CPU运行模式，适用于无GPU加速的场景。同时，优化了多语言支持，新增了日语、韩语、越南语等语言配置文件，并更新了README文档以反映新的运行模式和多语言支持。此外，修复了部分代码逻辑，提升了系统的稳定性和兼容性。
This commit is contained in:
Jason
2025-05-12 16:39:48 +08:00
parent 7049a24883
commit f78e985e1c
62 changed files with 5412 additions and 1520 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
docker/Dockerfile
View File

@@ -2,29 +2,53 @@ FROM python:3.12
RUN --mount=type=cache,target=/root/.cache,sharing=private \
apt update && \
apt install -y libgl1-mesa-glx && \
apt install -y libgl1-mesa-glx \
# pyside6
libegl1 libxkbcommon0 libdbus-1-3 && \
true
ADD . /vsr
ARG CUDA_VERSION=11.8
ARG USE_DIRECTML=0
ARG HARDWARD_ACCELERATOR="cuda"
# 如果是 CUDA 版本，执行 CUDA 特定设置
# 如果是 CUDA 12.x 版本，执行 CUDA 12.x 特定设置
RUN --mount=type=cache,target=/root/.cache,sharing=private \
if [ "${USE_DIRECTML:-0}" != "1" ]; then \
if [ "${HARDWARD_ACCELERATOR}" = "cuda" ] && [ "${CUDA_VERSION}" != "11.8" ]; then \
pip install paddlepaddle==3.0 && \
pip install torch==2.7.0 torchvision==0.22.0 --index-url https://download.pytorch.org/whl/cu$(echo ${CUDA_VERSION} | tr -d '.') && \
pip install -r /vsr/requirements.txt; \
fi
# 如果是 CUDA 11.8 版本，执行 CUDA 11.8 特定设置
RUN --mount=type=cache,target=/root/.cache,sharing=private \
if [ "${HARDWARD_ACCELERATOR}" = "cuda" ] && [ "${CUDA_VERSION}" = "11.8" ]; then \
pip install paddlepaddle==3.0 && \
pip install torch==2.3.1 torchvision==0.18.1 --index-url https://download.pytorch.org/whl/cu$(echo ${CUDA_VERSION} | tr -d '.') && \
pip install -r /vsr/requirements.txt && \
pip uninstall -y onnxruntime-gpu && \
pip install onnxruntime-gpu==1.20.1 --index-url https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/onnxruntime-cuda-11/pypi/simple/ && \
# for paddle
pip install setuptools==80.4.0; \
fi
# 如果是 DirectML 版本，执行 DirectML 特定设置
RUN --mount=type=cache,target=/root/.cache,sharing=private \
if [ "${USE_DIRECTML:-0}" = "1" ]; then \
if [ "${HARDWARD_ACCELERATOR}" = "directml" ]; then \
pip install paddlepaddle==3.0 && \
pip install torch_directml==0.2.5.dev240914 && \
pip install -r /vsr/requirements.txt; \
fi
# 如果是 CPU 版本，执行 CPU 特定设置
RUN --mount=type=cache,target=/root/.cache,sharing=private \
if [ "${HARDWARD_ACCELERATOR}" = "cpu" ]; then \
pip install paddlepaddle==3.0 && \
pip install -r /vsr/requirements.txt && \
sed -i 's/HARDWARD_ACCELERATION_OPTION *= *.*/HARDWARD_ACCELERATION_OPTION = False/g' /vsr/backend/config.py; \
fi
ENV LD_LIBRARY_PATH=/usr/local/lib/python3.12/site-packages/nvidia/cudnn/lib/
ENV LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib/python3.12/site-packages/nvidia/cuda_runtime/lib/
ENV LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib/python3.12/site-packages/nvidia/cuda_nvrtc/lib/
WORKDIR /vsr
CMD ["python", "/vsr/backend/main.py"]