Update app.py

app.py

@@ -148,7 +148,6 @@ def segment_frame(evt: gr.SelectData, label, video_state):
 
     return Image.fromarray(painted_image)
 
-@spaces.GPU(duration=10)
 def clear_clicks(video_state):
     video_state["input_points"] = []
     video_state["input_labels"] = []
@@ -180,6 +179,7 @@ def preprocess_for_removal(images, masks):
     arr_masks = np.stack(out_masks)
     return torch.from_numpy(arr_images).half().to(device), torch.from_numpy(arr_masks).half().to(device)
 
+@spaces.GPU(duration=200)
 def inference_and_return_video(dilation_iterations, num_inference_steps, video_state=None):
     if video_state["origin_images"] is None or video_state["masks"] is None:
         return None
@@ -218,6 +218,7 @@ def inference_and_return_video(dilation_iterations, num_inference_steps, video_s
     clip.write_videofile(video_file, codec='libx264', audio=False, verbose=False, logger=None)
     return video_file
 
+@spaces.GPU(duration=60)
 def track_video(n_frames, video_state):
 
     input_points = video_state["input_points"]