Skip OP/ED files by default (--all-files to include)

Real-run evidence: the transformation-emblem false positive (frame 1583
of the OP, 717x857, aspect 0.84) is geometrically face-shaped and
passes every box gate; OP/ED music videos are dominated by exactly that
kind of stylized graphics. Episodes carry the character close-ups a
grounding dataset needs. Verified on the second run: feather storm
(frame 29, area .64) now correctly rejected by the area cap.

README.md

@@ -64,6 +64,7 @@ context leaks onto the wrong card:
 | `--edge-margin` | 0 | require faces this far inside frame edges; `0` accepts edge-touching faces (original hard-rejected them) |
 | `--scene-threshold` | 30.0 | PySceneDetect ContentDetector threshold (same as original) |
 | `--imgsz` | 1280 | inference size; model was trained at 1280, `640` is faster at some recall cost |
+| `--all-files` | off | also scan OP/ED files. Off by default: music videos are full of stylized graphics (feather storms, transformation emblems) that false-positive as face-shaped; episodes carry the real character close-ups |
 | `--preview-dir` | off | also write annotated copies (green boxes) of every saved frame for eyeball QC |
 
 ## Notes

anime_facesaver.py

@@ -132,6 +132,15 @@ def parse_arguments():
         help="Inference image size. The model was trained at 1280; lower "
              "it (640) for speed at some recall cost.",
     )
+    parser.add_argument(
+        "--all-files",
+        action="store_true",
+        help="Also scan OP/ED files (name matches 'opening'/'ending'). "
+             "Skipped by default: music-video files are dominated by "
+             "stylized graphics (feather storms, transformation "
+             "emblems) that false-positive face detectors; character "
+             "close-ups for training datasets come from episodes.",
+    )
     parser.add_argument(
         "--preview-dir",
         default=None,
@@ -367,11 +376,18 @@ def main():
     video_extensions = ('.mp4', '.avi', '.mov', '.mkv')
     grand_total = 0
     for filename in sorted(os.listdir(args.input_dir)):
-        if filename.lower().endswith(video_extensions):
-            video_path = os.path.join(args.input_dir, filename)
-            print(f"Processing video: {video_path}")
-            grand_total += process_video(
-                video_path, args.output_dir, args, model, device, draw)
+        if not filename.lower().endswith(video_extensions):
+            continue
+        if (not args.all_files
+                and any(k in filename.lower()
+                        for k in ("opening", "ending", " op", " ed"))):
+            print(f"Skipping OP/ED file (use --all-files to include): "
+                  f"{filename}")
+            continue
+        video_path = os.path.join(args.input_dir, filename)
+        print(f"Processing video: {video_path}")
+        grand_total += process_video(
+            video_path, args.output_dir, args, model, device, draw)
 
     print(f"Done. {grand_total} stills saved to {args.output_dir}")