Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

what does this A_25 = np.array([[... means in HPGNet.py? #1

Open
southsea0725 opened this issue Apr 16, 2021 · 1 comment
Open

what does this A_25 = np.array([[... means in HPGNet.py? #1

southsea0725 opened this issue Apr 16, 2021 · 1 comment

Comments

@southsea0725
Copy link

Would you please tell me A_25 = np.array([[... means ?
If I use 17 key points, how to modify this variable ?
thanks

@niais
Copy link
Owner

niais commented Apr 16, 2021

Would you please tell me A_25 = np.array([[... means ?
If I use 17 key points, how to modify this variable ?
thanks

A_25 here is an embedding matrix that clusters the skeleton (25 joints) into 5 groups. Please refer to the paper for the skeleton topology.

# skeleton partitions for NTU-RGB+D
trunk: [1, 2, 3, 4]
left arm: [5, 6, 7, 8, 22, 23]
right arm: [9, 10, 11, 12, 24, 25]
left leg: [13, 14, 15, 16]
right leg: [17, 18, 19, 20]

We use it to embed 25 joints into 5 parts to extract higher-level semantic features.
BTW, you can simply uncomment A_embed = torch.ones([A.size(0), A.size(1), group_num]) in HPGNet to let the network learn clusters itself. This operation won't decrease the performance.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants