Fix

app.py

@@ -17,7 +17,7 @@ model = RTDetrForObjectDetection.from_pretrained("PekingU/rtdetr_r50vd")
 def stream_object_detection(video, conf_threshold):
     cap = cv2.VideoCapture(video)
 
-    video_codec = cv2.VideoWriter_fourcc(*"x264") # type: ignore
+    video_codec = cv2.VideoWriter_fourcc(*"mp4v") # type: ignore
     fps = int(cap.get(cv2.CAP_PROP_FPS))
     desired_fps = fps // 5
     width  = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
@@ -28,11 +28,12 @@ def stream_object_detection(video, conf_threshold):
     n_frames = 0
     n_chunks = 0
 
-    name = f"output_{n_chunks}.ts"
-    segment_file = cv2.VideoWriter(name, video_codec, fps, (width, height)) # type: ignore
+    name = f"output_{n_chunks}.mp4"
+    segment_file = cv2.VideoWriter(name, video_codec, desired_fps, (width // 2, height // 2)) # type: ignore
     batch = []
 
     while iterating:
+        frame = cv2.resize( frame, (0,0), fx=0.5, fy=0.5)
         frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
         if n_frames % 5 == 0:
             batch.append(frame)
@@ -62,7 +63,7 @@ def stream_object_detection(video, conf_threshold):
             n_frames = 0
             n_chunks += 1
             yield name
-            name = f"output_{n_chunks}.ts"
+            name = f"output_{n_chunks}.mp4"
             segment_file = cv2.VideoWriter(name, video_codec, fps, (width, height)) # type: ignore
 
         iterating, frame = cap.read()
@@ -91,7 +92,7 @@ with gr.Blocks(css=css) as app:
         """)
     with gr.Column(elem_classes=["my-column"]):
         with gr.Group(elem_classes=["my-group"]):
-            video = gr.Video(label="Video Source", streaming=True, autoplay=True)
+            video = gr.Video(label="Video Source", streaming=True)
             conf_threshold = gr.Slider(
                 label="Confidence Threshold",
                 minimum=0.0,