20260430_001744-feat: PyTorch CUDA + SAM2 GPU inference, video thumbnail, real FPS + configurable parse FPS, DICOM batch import

backend/services/frame_parser.py

@@ -5,7 +5,7 @@ import os
 import shutil
 import subprocess
 from pathlib import Path
-from typing import List, Optional
+from typing import List, Optional, Tuple
 
 import cv2
 import numpy as np
@@ -16,12 +16,43 @@ from minio_client import upload_file, BUCKET_NAME
 logger = logging.getLogger(__name__)
 
 
+def get_video_fps(video_path: str) -> float:
+    """Read the original frame rate of a video file."""
+    cap = cv2.VideoCapture(video_path)
+    if not cap.isOpened():
+        return 30.0
+    fps = cap.get(cv2.CAP_PROP_FPS)
+    cap.release()
+    return fps if fps > 0 else 30.0
+
+
+def extract_thumbnail(video_path: str, output_path: str, width: int = 640) -> str:
+    """Extract the first frame of a video as a thumbnail JPEG."""
+    cap = cv2.VideoCapture(video_path)
+    if not cap.isOpened():
+        raise RuntimeError(f"Cannot open video for thumbnail: {video_path}")
+    ret, frame = cap.read()
+    cap.release()
+    if not ret or frame is None:
+        raise RuntimeError(f"Cannot read first frame from: {video_path}")
+
+    h, w = frame.shape[:2]
+    if w > width:
+        scale = width / w
+        new_w = int(w * scale)
+        new_h = int(h * scale)
+        frame = cv2.resize(frame, (new_w, new_h), interpolation=cv2.INTER_AREA)
+
+    cv2.imwrite(output_path, frame, [cv2.IMWRITE_JPEG_QUALITY, 85])
+    return output_path
+
+
 def parse_video(
     video_path: str,
     output_dir: str,
     fps: int = 30,
     max_frames: Optional[int] = None,
-) -> List[str]:
+) -> Tuple[List[str], float]:
     """Extract frames from a video file using FFmpeg or OpenCV fallback.
 
     Args:
@@ -31,10 +62,11 @@ def parse_video(
         max_frames: Optional maximum number of frames to extract.
 
     Returns:
-        List of paths to extracted frame images.
+        Tuple of (frame_paths, original_fps).
     """
     os.makedirs(output_dir, exist_ok=True)
     frame_paths: List[str] = []
+    original_fps = get_video_fps(video_path)
 
     # Try FFmpeg first
     if shutil.which("ffmpeg"):
@@ -57,7 +89,7 @@ def parse_video(
                 if max_frames:
                     frame_paths = frame_paths[:max_frames]
                 logger.info("Extracted %d frames via FFmpeg", len(frame_paths))
-                return frame_paths
+                return frame_paths, original_fps
             else:
                 logger.warning("FFmpeg failed: %s", result.stderr)
         except Exception as exc:  # noqa: BLE001
@@ -89,7 +121,7 @@ def parse_video(
 
     cap.release()
     logger.info("Extracted %d frames via OpenCV", len(frame_paths))
-    return frame_paths
+    return frame_paths, original_fps
 
 
 def parse_dicom(
@@ -134,12 +166,12 @@ def parse_dicom(
             # Handle multi-frame DICOM
             if pixel_array.ndim == 3:
                 for f in range(pixel_array.shape[0]):
-                    out_path = os.path.join(output_dir, f"frame_{idx:06d}_{f:03d}.png")
-                    cv2.imwrite(out_path, pixel_array[f])
+                    out_path = os.path.join(output_dir, f"frame_{idx:06d}_{f:03d}.jpg")
+                    cv2.imwrite(out_path, pixel_array[f], [cv2.IMWRITE_JPEG_QUALITY, 85])
                     frame_paths.append(out_path)
             else:
-                out_path = os.path.join(output_dir, f"frame_{idx:06d}.png")
-                cv2.imwrite(out_path, pixel_array)
+                out_path = os.path.join(output_dir, f"frame_{idx:06d}.jpg")
+                cv2.imwrite(out_path, pixel_array, [cv2.IMWRITE_JPEG_QUALITY, 85])
                 frame_paths.append(out_path)
         except Exception as exc:  # noqa: BLE001
             logger.error("Failed to read DICOM %s: %s", path, exc)