search

Jho-Yonsei / HD-GCN

[ICCV 2023] Hierarchically Decomposed Graph Convolutional Networks for Skeleton-Based Action Recognition
MIT License
131 stars 23 forks source link

May I ask if it is possible to upload the file for Generate NTU RGB+D 60 or NTU RGB+D 120 dataset？ #9

Open ikbears opened 1 year ago

ikbears commented 1 year ago

May I ask if it is possible to upload the file for Generate NTU RGB+D 60 or NTU RGB+D 120 dataset？

Jho-Yonsei commented 1 year ago

I will upload those codes as soon as possible.

If you need those codes now, you can refer CTR-GCN github page.

Thank you!

ikbears commented 1 year ago

I will upload those codes as soon as possible.

If you need those codes now, you can refer CTR-GCN github page.

Thank you!

Thank you！Best wish！

ikbears commented 1 year ago

I will upload those codes as soon as possible.

If you need those codes now, you can refer CTR-GCN github page.

Thank you!

Hello, I have replicated your experiments on the NTU 60 and 120 datasets. I would like to replicate the experiments on the NW-UCLA dataset now, but I found that there is no config file for the NW-UCLA dataset in the warehouse. Can you upload it? Thank you.

cindyma919 commented 2 months ago

I will upload those codes as soon as possible. If you need those codes now, you can refer CTR-GCN github page. Thank you!

Hello, I have replicated your experiments on the NTU 60 and 120 datasets. I would like to replicate the experiments on the NW-UCLA dataset now, but I found that there is no config file for the NW-UCLA dataset in the warehouse. Can you upload it? Thank you.

Hello, I also replicated the experiment on the NTU 60 and 120 datasets. But the accuracy of my replication is not as high as the accuracy in the article. I would like to ask if you have achieved the accuracy mentioned in the article? I suspect that this issue is caused by different environmental configurations. So may I ask if you are willing to share the version information of your Python, PyTorch, CUDA, and CuDNN? Your sharing will be of great help to me. Thank you！