Update app.py

app.py

@@ -53,41 +53,43 @@ def process_input(input_file):
 
     if ext in ['.mp4', '.avi', '.mov']:
         cap = cv2.VideoCapture(file_path)
-
         if not cap.isOpened():
-            return None, "Could not open video file", ""
+            return None, None, "Could not open video file", ""
 
         fps = cap.get(cv2.CAP_PROP_FPS)
-        width = 640
-        height = 640
-
-        # Generate safe output path
-        output_path = f"annotated_output.mp4"
-        fourcc = cv2.VideoWriter_fourcc(*'mp4v')
-        out = cv2.VideoWriter(output_path, fourcc, fps, (width, height))
+        frame_skip = 5
+        frame_id = 0
+        output_path = "annotated_output.mp4"
+        out = cv2.VideoWriter(output_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (640, 640))
 
         while True:
             ret, frame = cap.read()
             if not ret:
                 break
-            frame_resized = cv2.resize(frame, (640, 640))
-            annotated_frame = annotate_frame(frame_resized)
-            bgr_frame = cv2.cvtColor(annotated_frame, cv2.COLOR_RGB2BGR)
-            out.write(bgr_frame)
+            if frame_id % frame_skip != 0:
+                frame_id += 1
+                continue
+
+            frame_id += 1
+            frame = cv2.resize(frame, (640, 640))
+            annotated = annotate_frame(frame)
+            out.write(cv2.cvtColor(annotated, cv2.COLOR_RGB2BGR))
 
         cap.release()
         out.release()
 
-        return output_path, "Bangla text in video (see frames)", "OCR confidence displayed on frames"
+        return output_path, None, "Bangla text in video (see frames)", "OCR confidence displayed"
 
     else:
-        # Handle image
         frame = cv2.imread(file_path)
         if frame is None:
-            return None, "Invalid image file", ""
+            return None, None, "Invalid image", ""
 
+        frame = cv2.resize(frame, (640, 640))
         annotated = annotate_frame(frame)
-        return Image.fromarray(annotated), "Bangla text in image", ""
+        pil_img = Image.fromarray(annotated)
+        return None, pil_img, "Bangla text in image", "OCR confidence in image"
+
 
 
 interface = gr.Interface(