[Docs] how to visualization of the result of posec3d

[jiaqihuang01]

The doc issue

Thank you for your excellent paper. When using posec3d, I first converted the video into a pkl file, and then performed training and testing with better results. My confusion is how to visualize the posec3d recognition effect。 I tried to modify the show settings in the test .py variable, and added visualization=dict(type="VisualizationHook") in my config, and an error like this appeared：

File "c:\study\mmaction2-main\mmaction\ engine\hooks\visualization_hook.py", line 81, in _draw_samples
     video = video.permute(1, 2, 3, 0).numpy().astype('uint8')
RuntimeError: permute(sparse_coo): number of dimensions in the tensor input does not match the length of the desired ordering of dimensions i.e. input.dim() = 5 is not equal to len(dims) = 4. 

In posec3d, I only imported the pkl file. How can I visualize the video results well?

Suggest a potential alternative/fix

No response