In regress_kp(bctch_list) from celeba_wild.py, there are only 2 keys 'det_keypoints' and 'keypoints'.
Should I add "'eye_distance': batch['eye_distance'].cpu()" to line 93 of model.py which is originally "out_batch = {'keypoints': batch['keypoints'].cpu(), 'det_keypoints': kp.cpu()}"?
I think your CVPR 2023 paper FewShot3DKP (Few-shot Geometry-Aware Keypoint Localization) is very insightful for my current work. Will the code of this work be publicly available?
thanks to this nice work, I try to train autoLink on wflw with five keypoints because of low cuda memory. but I don't get the desired result. Any advice is helpful,this is my result.
Thank you very much for making such great work public!
I couldn't find list_imsize_celeba.txt and celebA_training.txt from the CelebA Dataset website , can you provide more instructions on how to obtain these two files? Thank you very much for your help.
I have downloaded the dataset from your recommended website. But when i use the preprocess code to generate '.h5' file for cub_200_2011 dataset, I cant find the file '/cub/CUB_200_2011/cachedir/cub/data/train_cub_cleaned.mat'.