Open zhouxy-1999 opened 1 year ago
I sincerely apologize for the late reply. The npy files are obtained from the json files, which record the keypoints annotation. I'm sorry to say that I forgot to correct this part of code when pushing the repo. (would be corrected soon) You could simply process the json into npy or modify the code which load the keypoints.
When I was running train_recon.py, the following error occurred:
I failed to solve the problem,and I hope for some help.