继续修复bug

2026-02-04 04:34:41 +08:00 · 2023-12-27 20:32:00 +08:00
parent f92a483717
commit 313c3d37a7
4 changed files with 74 additions and 37 deletions
--- a/backend/config.py
+++ b/backend/config.py
@@ -20,32 +20,34 @@ DET_MODEL_BASE = os.path.join(BASE_DIR, 'models')
 DET_MODEL_PATH = os.path.join(DET_MODEL_BASE, MODEL_VERSION, 'ch_det')

 # ×××××××××××××××××××× [可以改] start ××××××××××××××××××××
+# 是否使用全局mask
+SKIP_DETECTION = False
 # 单个字符的高度大于宽度阈值
 HEIGHT_WIDTH_DIFFERENCE_THRESHOLD = 10
 # 容忍的像素点偏差
 PIXEL_TOLERANCE_Y = 20  # 允许检测框纵向偏差50个像素点
 PIXEL_TOLERANCE_X = 20  # 允许检测框横向偏差100个像素点
 # 字幕区域偏移量， 放大诗歌像素点，防止字幕偏移
-SUBTITLE_AREA_DEVIATION_PIXEL = 10
+SUBTITLE_AREA_DEVIATION_PIXEL = 20
 # 20个像素点以内的差距认为是同一行
 TOLERANCE_Y = 20
 # 高度差阈值
 THRESHOLD_HEIGHT_DIFFERENCE = 20
-# 相邻帧出
+# 相邻帧数
 NEIGHBOR_STRIDE = 5
 # 参考帧长度
 REFERENCE_LENGTH = 5
+# 模式列表，请根据自己需求选择inpaint模式
+# ACCURATE模式将消耗大量GPU显存，如果您的显卡显存较少，建议设置为NORMAL
+MODE_LIST = ['FAST', 'NORMAL', 'ACCURATE']
+MODE = 'NORMAL'
 # 【根据自己的GPU显存大小设置】最大同时处理的图片数量，设置越大处理效果越好，但是要求显存越高
 # 1280x720p视频设置80需要25G显存，设置50需要19G显存
 # 720x480p视频设置80需要8G显存，设置50需要7G显存
 MAX_PROCESS_NUM = 70
 # 【根据自己内存大小设置】设置的越大效果越好，但是时间越长
 MAX_LOAD_NUM = 20
-# 模式列表，请根据自己需求选择inpiant模式
-# ACCURATE模式将消耗大量GPU显存，如果您的显卡显存较少，建议设置为NORMAL
-MODE_LIST = ['FAST', 'NORMAL', 'ACCURATE']
-MODE = 'NORMAL'
-# 如果仅需要去除文字区域，则使用FAST
+# 如果仅需要去除文字区域，则可以将SUPER_FAST设置为True
 SUPER_FAST = False
 # ×××××××××××××××××××× [可以改] start ××××××××××××××××××××

@@ -83,4 +85,6 @@ os.chmod(FFMPEG_PATH, stat.S_IRWXU+stat.S_IRWXG+stat.S_IRWXO)
 os.environ['KMP_DUPLICATE_LIB_OK'] = 'True'
 if SUPER_FAST:
    MODE = 'FAST'
+if SKIP_DETECTION:
+    MODE = 'NORMAL'
 # ×××××××××××××××××××× [不要改] end ××××××××××××××××××××
--- a/backend/inpaint/sttn_inpaint.py
+++ b/backend/inpaint/sttn_inpaint.py
@@ -192,7 +192,10 @@ class STTNInpaint:
                        to_H += move
                        from_H += move
                # 将该段落添加到列表中
-                inpaint_area.append((from_H, to_H))
+                if (from_H, to_H) not in inpaint_area:
+                    inpaint_area.append((from_H, to_H))
+                else:
+                    break
            # 移动到下一个段落
            to_H -= h
        return inpaint_area  # 返回绘画区域列表
@@ -210,15 +213,8 @@ class STTNVideoInpaint:
            'fps': reader.get(cv2.CAP_PROP_FPS),  # 视频的帧率
            'len': int(reader.get(cv2.CAP_PROP_FRAME_COUNT) + 0.5)  # 视频的总帧数
        }
-        # 创建视频写入对象，用于输出修复后的视频
-        writer = cv2.VideoWriter(
-            self.video_out_path,
-            cv2.VideoWriter_fourcc(*"mp4v"),
-            frame_info['fps'],
-            (frame_info['W_ori'], frame_info['H_ori'])
-        )
        # 返回视频读取对象、帧信息和视频写入对象
-        return reader, frame_info, writer
+        return reader, frame_info

    def __init__(self, video_path, mask_path=None, clip_gap=None):
        # STTNInpaint视频修复实例初始化
@@ -237,16 +233,24 @@ class STTNVideoInpaint:
        else:
            self.clip_gap = clip_gap

-    def __call__(self, mask=None):
+    def __call__(self, input_mask=None, input_video_writer=None):
        # 读取视频帧信息
-        reader, frame_info, writer = self.read_frame_info_from_video()
+        reader, frame_info = self.read_frame_info_from_video()
+        if input_video_writer is not None:
+            writer = input_video_writer
+        else:
+            # 创建视频写入对象，用于输出修复后的视频
+            writer = cv2.VideoWriter(self.video_out_path, cv2.VideoWriter_fourcc(*"mp4v"), frame_info['fps'], (frame_info['W_ori'], frame_info['H_ori']))
        # 计算需要迭代修复视频的次数
        rec_time = frame_info['len'] // self.clip_gap if frame_info['len'] % self.clip_gap == 0 else frame_info['len'] // self.clip_gap + 1
        # 计算分割高度，用于确定修复区域的大小
        split_h = int(frame_info['W_ori'] * 3 / 16)
-        if mask is None:
+        if input_mask is None:
            # 读取掩码
            mask = self.sttn_inpaint.read_mask(self.mask_path)
+        else:
+            _, mask = cv2.threshold(input_mask, 127, 1, cv2.THRESH_BINARY)
+            mask = mask[:, :, None]
        # 得到修复区域位置
        inpaint_area = self.sttn_inpaint.get_inpaint_area_by_mask(frame_info['H_ori'], split_h, mask)
        # 遍历每一次的迭代次数
--- a/backend/main.py
+++ b/backend/main.py
@@ -5,12 +5,15 @@ from pathlib import Path
 import threading
 import cv2
 import sys
+
+import numpy as np
+
 sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 import config
 from backend.scenedetect import scene_detect
 from backend.scenedetect.detectors import ContentDetector
-from backend.inpaint.sttn_inpaint import STTNInpaint
+from backend.inpaint.sttn_inpaint import STTNInpaint, STTNVideoInpaint
 from backend.inpaint.lama_inpaint import LamaInpaint
 from backend.inpaint.video_inpaint import VideoInpaint
 from backend.tools.inpaint_tools import create_mask, batch_generator
@@ -567,8 +570,10 @@ class SubtitleRemover:
        self.progress_total = 50 + self.progress_remover

    def propainter_mode(self, sub_list, continuous_frame_no_list, tbar):
-        # *********************** 批推理方案 start ***********************
        print('use propainter mode')
+        scene_div_points = self.sub_detector.get_scene_div_frame_no(self.video_path)
+        continuous_frame_no_list = self.sub_detector.split_range_by_scene(continuous_frame_no_list,
+                                                                          scene_div_points)
        self.video_inpaint = VideoInpaint(config.MAX_PROCESS_NUM)
        index = 0
        while True:
@@ -647,7 +652,20 @@ class SubtitleRemover:
                                        if self.gui_mode:
                                            self.preview_frame = cv2.hconcat([batch[i], inpainted_frame])
                                self.update_progress(tbar, increment=len(batch))
-        # *********************** 批推理方案 end ***********************
+
+    def sttn_mode_with_no_detection(self):
+        """
+        选中区域，不进行字幕检测
+        """
+        print('use sttn mode with no detection')
+        if self.sub_area is not None:
+            ymin, ymax, xmin, xmax = self.sub_area
+            mask_area_coordinates = [(xmin, xmax, ymin, ymax)]
+            mask = create_mask(self.mask_size, mask_area_coordinates)
+            sttn_video_inpaint = STTNVideoInpaint(self.video_path)
+            sttn_video_inpaint(input_mask=mask, input_video_writer=self.video_writer)
+        else:
+            print('please set subtitle area first')

    def sttn_mode(self, sub_list, continuous_frame_no_list, tbar):
        # *********************** 批推理方案 start ***********************
@@ -747,17 +765,10 @@ class SubtitleRemover:
        start_time = time.time()
        # 重置进度条
        self.progress_total = 0
-        # 寻找字幕帧
-        sub_list = self.sub_detector.find_subtitle_frame_no(sub_remover=self)
-        continuous_frame_no_list = self.sub_detector.find_continuous_ranges_with_same_mask(sub_list)
-        # 获取场景分割的帧号
-        scene_div_points = self.sub_detector.get_scene_div_frame_no(self.video_path)
-        continuous_frame_no_list = self.sub_detector.split_range_by_scene(continuous_frame_no_list, scene_div_points)
        tbar = tqdm(total=int(self.frame_count), unit='frame', position=0, file=sys.__stdout__,
                    desc='Subtitle Removing')
-        print('[Processing] start removing subtitles...')
-
        if self.is_picture:
+            sub_list = self.sub_detector.find_subtitle_frame_no(sub_remover=self)
            self.lama_inpaint = LamaInpaint()
            original_frame = cv2.imread(self.video_path)
            mask = create_mask(original_frame.shape[0:2], sub_list[1])
@@ -768,12 +779,22 @@ class SubtitleRemover:
            tbar.update(1)
            self.progress_total = 100
        else:
-            if config.MODE == 'ACCURATE':
-                self.propainter_mode(sub_list, continuous_frame_no_list, tbar)
-            elif config.MODE == 'NORMAL':
-                self.sttn_mode(sub_list, continuous_frame_no_list, tbar)
+            # 是否跳过字幕帧寻找
+            if config.SKIP_DETECTION:
+                # 若跳过则世界使用sttn模式
+                print('[Processing] start removing subtitles...')
+                self.sttn_mode_with_no_detection()
            else:
-                self.lama_mode(sub_list, tbar)
+                sub_list = self.sub_detector.find_subtitle_frame_no(sub_remover=self)
+                continuous_frame_no_list = self.sub_detector.find_continuous_ranges_with_same_mask(sub_list)
+                print('[Processing] start removing subtitles...')
+                # 精准模式下，获取场景分割的帧号，进一步切割
+                if config.MODE == 'ACCURATE':
+                    self.propainter_mode(sub_list, continuous_frame_no_list, tbar)
+                elif config.MODE == 'NORMAL':
+                    self.sttn_mode(sub_list, continuous_frame_no_list, tbar)
+                else:
+                    self.lama_mode(sub_list, tbar)
        self.video_cap.release()
        self.video_writer.release()
        if not self.is_picture:
--- a/backend/tools/inpaint_tools.py
+++ b/backend/tools/inpaint_tools.py
@@ -78,8 +78,16 @@ def create_mask(size, coords_list):
        for coords in coords_list:
            xmin, xmax, ymin, ymax = coords
            # 为了避免框过小，放大10个像素
-            cv2.rectangle(mask, (xmin - config.SUBTITLE_AREA_DEVIATION_PIXEL, ymin - config.SUBTITLE_AREA_DEVIATION_PIXEL),
-                          (xmax + config.SUBTITLE_AREA_DEVIATION_PIXEL, ymax + config.SUBTITLE_AREA_DEVIATION_PIXEL), (255, 255, 255), thickness=-1)
+            x1 = xmin - config.SUBTITLE_AREA_DEVIATION_PIXEL
+            if x1 < 0:
+                x1 = 0
+            y1 = ymin - config.SUBTITLE_AREA_DEVIATION_PIXEL
+            if y1 < 0:
+                y1 = 0
+            x2 = xmax + config.SUBTITLE_AREA_DEVIATION_PIXEL
+            y2 = ymax + config.SUBTITLE_AREA_DEVIATION_PIXEL
+            cv2.rectangle(mask, (x1, y1),
+                          (x2, y2), (255, 255, 255), thickness=-1)
    return mask