Upload 3 files

README.md

@@ -4,7 +4,7 @@ emoji: 🎬
 colorFrom: blue
 colorTo: purple
 sdk: gradio
-sdk_version: 4.0.0
+sdk_version: 4.44.0
 app_file: app.py
 pinned: false
 license: apache-2.0

app.py

@@ -50,9 +50,20 @@ def extract_frames_from_video(video_path, max_frames=30):
     timestamps = []
     
     try:
+        print(f"Attempting to extract frames from: {video_path}")
         cap = cv2.VideoCapture(video_path)
+        
+        if not cap.isOpened():
+            print(f"Failed to open video: {video_path}")
+            return [], []
+            
         fps = cap.get(cv2.CAP_PROP_FPS)
-        frame_interval = int(fps)  # Extract 1 frame per second
+        total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
+        duration = total_frames / fps if fps > 0 else 0
+        
+        print(f"Video info - FPS: {fps}, Total frames: {total_frames}, Duration: {duration:.2f}s")
+        
+        frame_interval = max(1, int(fps))  # Extract 1 frame per second, minimum 1
         
         frame_count = 0
         extracted_count = 0
@@ -68,10 +79,12 @@ def extract_frames_from_video(video_path, max_frames=30):
                 frames.append(Image.fromarray(frame_rgb))
                 timestamps.append(extracted_count)
                 extracted_count += 1
+                print(f"Extracted frame {extracted_count} at {frame_count}/{total_frames}")
             
             frame_count += 1
         
         cap.release()
+        print(f"Successfully extracted {len(frames)} frames")
         return frames, timestamps
     except Exception as e:
         print(f"Error extracting frames: {e}")
@@ -226,16 +239,30 @@ def process_video_with_minicpm(video_file):
     try:
         start_time = time.time()
         
+        # Handle both file object and string path
+        if hasattr(video_file, 'name'):
+            video_path = video_file.name
+        else:
+            video_path = video_file
+        
+        # Debug: Check what we received
+        print(f"Video input type: {type(video_file)}")
+        print(f"Video path: {video_path}")
+        
+        # Validate file exists
+        if not os.path.exists(video_path):
+            return f"Video file not found: {video_path}", "", ""
+        
         # Extract frames
         update_status = "Extracting frames from video..."
-        frames, timestamps = extract_frames_from_video(video_file.name)
+        frames, timestamps = extract_frames_from_video(video_path)
         
         if not frames:
             return "Failed to extract frames from video.", "", ""
         
         # Extract audio
         update_status = "Extracting audio from video..."
-        audio_path = extract_audio_from_video(video_file.name)
+        audio_path = extract_audio_from_video(video_path)
         
         # Analyze with MiniCPM-o
         update_status = "Analyzing content with MiniCPM-o..."

requirements.txt

@@ -1,6 +1,6 @@
 torch>=2.1.0
 transformers>=4.35.0
-gradio>=4.0.0
+gradio>=4.44.0
 opencv-python>=4.8.0
 numpy>=1.24.0
 pillow>=10.0.0