Spaces:

jbilcke-hf

/

FacePoke

like 1.6k

Running on L40S

Traceback (most recent call last):
  File "/home/user/app/app.py", line 76, in websocket_handler
    res = await engine.load_image(msg.data)
  File "/home/user/.local/lib/python3.10/site-packages/async_lru/__init__.py", line 294, in __call__
    return await self.__wrapper(self.__instance, *fn_args, **fn_kwargs)
  File "/home/user/.local/lib/python3.10/site-packages/async_lru/__init__.py", line 227, in __call__
    return await asyncio.shield(fut)
  File "/home/user/app/engine.py", line 73, in load_image
    crop_info = await asyncio.to_thread(self.live_portrait.cropper.crop_single_image, img_rgb)
  File "/usr/lib/python3.10/asyncio/threads.py", line 25, in to_thread
    return await loop.run_in_executor(None, func_call)
  File "/usr/lib/python3.10/concurrent/futures/thread.py", line 58, in run
    result = self.fn(*self.args, **self.kwargs)
  File "/home/user/app/liveportrait/utils/cropper.py", line 73, in crop_single_image
    src_face = self.face_analysis_wrapper.get(
  File "/home/user/app/liveportrait/utils/face_analysis_diy.py", line 47, in get
    bboxes, kpss = self.det_model.detect(img_bgr, max_num=max_num, metric='default')
  File "/home/user/app/liveportrait/utils/dependencies/insightface/model_zoo/retinaface.py", line 222, in detect
    det_img[:new_height, :new_width, :] = resized_img
ValueError: could not broadcast input array from shape (352,512,4) into shape (352,512,3)

You could fix this by removing the alpha channel from the image