Gradio App giving NaN error on WIndows 11

[naseemi9]

Hi , It used to work fine when I executed run_windows_human.bat file but suddenly I have started receiving NaN error when I click on Animate button (ie. start processing).

I have Windows 11 with Nvidia T500 (as I said , It used to work as expected , I also tried to delete contents of %TEMP%)

Here is the output:

The animated video consists of 78 frames.                                       live_portrait_pipeline.py:270
🚀Animating... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━   0% -:--:--
Traceback (most recent call last):
  File "D:\NSM3\LivePortrait\LivePortrait-Org\LivePortrait_env\lib\site-packages\gradio\queueing.py", line 541, in process_events
    response = await route_utils.call_process_api(
  File "D:\NSM3\LivePortrait\LivePortrait-Org\LivePortrait_env\lib\site-packages\gradio\route_utils.py", line 276, in call_process_api
    output = await app.get_blocks().process_api(
  File "D:\NSM3\LivePortrait\LivePortrait-Org\LivePortrait_env\lib\site-packages\gradio\blocks.py", line 1928, in process_api
    result = await self.call_function(
  File "D:\NSM3\LivePortrait\LivePortrait-Org\LivePortrait_env\lib\site-packages\gradio\blocks.py", line 1514, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "D:\NSM3\LivePortrait\LivePortrait-Org\LivePortrait_env\lib\site-packages\anyio\to_thread.py", line 56, in run_sync
    return await get_async_backend().run_sync_in_worker_thread(
  File "D:\NSM3\LivePortrait\LivePortrait-Org\LivePortrait_env\lib\site-packages\anyio\_backends\_asyncio.py", line 2177, in run_sync_in_worker_thread
    return await future
  File "D:\NSM3\LivePortrait\LivePortrait-Org\LivePortrait_env\lib\site-packages\anyio\_backends\_asyncio.py", line 859, in run
    result = context.run(func, *args)
  File "D:\NSM3\LivePortrait\LivePortrait-Org\LivePortrait_env\lib\site-packages\gradio\utils.py", line 833, in wrapper
    response = f(*args, **kwargs)
  File "D:\NSM3\LivePortrait\LivePortrait-Org\app.py", line 60, in gpu_wrapped_execute_video
    return gradio_pipeline.execute_video(*args, **kwargs)
  File "D:\NSM3\LivePortrait\LivePortrait-Org\LivePortrait_env\lib\site-packages\torch\autograd\grad_mode.py", line 28, in decorate_context
    return func(*args, **kwargs)
  File "D:\NSM3\LivePortrait\LivePortrait-Org\src\gradio_pipeline.py", line 219, in execute_video
    output_path, output_path_concat = self.execute(self.args)
  File "D:\NSM3\LivePortrait\LivePortrait-Org\src\live_portrait_pipeline.py", line 392, in execute
    motion_multiplier = calc_motion_multiplier(x_s, x_d_0_new)
  File "D:\NSM3\LivePortrait\LivePortrait-Org\src\utils\helper.py", line 37, in calc_motion_multiplier
    source_area = ConvexHull(kp_source_np.squeeze(0)).volume
  File "_qhull.pyx", line 2425, in scipy.spatial._qhull.ConvexHull.__init__
  File "_qhull.pyx", line 271, in scipy.spatial._qhull._Qhull.__init__
ValueError: Points cannot contain NaN

It seems to do something with ConvexHull

P.S. I used video and images given in examples of LivePortrait (this rules out the format/encoding issues)

Kindly help me in resolving the issue.

Thanks
Naseem