You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Could you tell me how to evaluate the model to get the metrics reported in the paper ?
In addition, I notice that in the codes of EHF dataset, you use a file named 'gt_keyps.npz'. But I did not find it on the website of EHF dataset. Could you tell me how to get this file?
Thanks!
The text was updated successfully, but these errors were encountered:
I have encountered the same problems. And I find there are some 2Djnt.json files in the EHF dataset, but contains (137,3) 2d keypoints information, without 3d keypoints and joints14 which are required in the ehf.py code. In addition, the ehf.py code read (144,2) 2d keypoints, not consistent with the content in 2Djnt.json files. So I'm wondering how to get these ground truth information of EHF.
I'm really eager to get some kind help! Thanks.
Hi,
Could you tell me how to evaluate the model to get the metrics reported in the paper ?
In addition, I notice that in the codes of EHF dataset, you use a file named 'gt_keyps.npz'. But I did not find it on the website of EHF dataset. Could you tell me how to get this file?
Thanks!
The text was updated successfully, but these errors were encountered: