Update app.py

app.py

@@ -69,20 +69,23 @@ def predict2(image_np):
     return result_pil_img
 
 def detect_video(video):
-    # Create a video capture object
-    cap = cv2.VideoCapture(video)
-
-    nb_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
-    frame_h = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
-    frame_w = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
-    fps = cap.get(cv2.CAP_PROP_FPS)
+    video_reader = cv2.VideoCapture(video)
+    
+    nb_frames = int(video_reader.get(cv2.CAP_PROP_FRAME_COUNT))
+    frame_h = int(video_reader.get(cv2.CAP_PROP_FRAME_HEIGHT))
+    frame_w = int(video_reader.get(cv2.CAP_PROP_FRAME_WIDTH))
+    fps = video_reader.get(cv2.CAP_PROP_FPS)
     
-    # Process frames in a loop
+    video_writer = cv2.VideoWriter(video_out_filepath,
+                               cv2.VideoWriter_fourcc(*'mp4v'), 
+                               fps, 
+                               (frame_w, frame_h))
+
     for i in tqdm(range(nb_frames)):
-        ret, image_np = cap.read()
+        ret, image_np = video_reader.read()
         input_tensor = tf.convert_to_tensor(np.expand_dims(image_np, 0), dtype=tf.uint8)
         results = detection_model(input_tensor)
-        image_np_with_detections = viz_utils.visualize_boxes_and_labels_on_image_array(
+        viz_utils.visualize_boxes_and_labels_on_image_array(
                   image_np,
                   results['detection_boxes'][0].numpy(),
                   (results['detection_classes'][0].numpy()+ label_id_offset).astype(int),
@@ -94,21 +97,14 @@ def detect_video(video):
                   agnostic_mode=False,
                   line_thickness=2)
 
+        video_writer.write(np.uint8(image_np))
+                
+    # Release camera and close windows
+    video_reader.release()
+    video_writer.release() 
+    cv2.destroyAllWindows() 
+    cv2.waitKey(1)
 
-        # Yield the processed frame
-        yield image_np_with_detections
-
-    # Release resources
-    cap.release()
-
- 
-inputs_video = [
-    gr.components.Video( label="Input Video"),
- 
-]
-outputs_video = [
-    gr.components.Image( label="Output Image"),
-]
 
 label_id_offset = 0    
 REPO_ID = "apailang/mytfodmodel"
@@ -142,25 +138,19 @@ tts_demo = gr.Interface(
     cache_examples=True
     )#.launch(share=True)
 
+samples_folder = 'data'
 
 a = os.path.join(os.path.dirname(__file__), "data/a.mp4")  # Video
 b = os.path.join(os.path.dirname(__file__), "data/b.mp4")  # Video
 c = os.path.join(os.path.dirname(__file__), "data/c.mp4")  # Video
 
 
-# interface_video = gr.Interface(
-#     fn=show_preds_video,
-#     inputs=inputs_video,
-#     outputs=outputs_video,
-#     title="Pothole detector",
-#     examples=video_path,
-#     cache_examples=False,
-# )
+video_out_file = os.path.join(samples_folder,'detected' + '.mp4')
 
 stt_demo = gr.Interface(
     fn=detect_video,
     inputs=inputs_video,
-    outputs=gr.Video(),
+    outputs="data/detected.mp4",
     examples=[
         [a],
         [b],