search

ChanglongJiangGit / A2J-Transformer

[CVPR 2023] Code for paper 'A2J-Transformer: Anchor-to-Joint Transformer Network for 3D Interacting Hand Pose Estimation from a Single RGB Image'
Apache License 2.0
85 stars 7 forks source link

visualize hand skeleton #22

Open qwer904 opened 5 months ago

qwer904 commented 5 months ago

Hello, I want to visualize the hand skeleton in the output picture, how should I do it? Are you using your vis_keypoints(img, kps, kps_gt, bbox, score, skeleton, filename, score_thr=0.4, line_width=3, circle_rad = 3, save_path=None) function? If so, how should the parameters img, kps, kps_gt, bbox, score, and skeleton be set?

ChanglongJiangGit commented 4 months ago

Sorry to keep you waiting.

Yes, if you want to visualize, just use this function. This is copied from the baseline (of InterHand2.6M) code.

Each parameter represents: img: input single image (before cropping); kps: your predicted coordinates of hand joints; kps_gt: GT coordinates of hand joints; bbox: GT bounding box; score: the joint_valid parameter which represents whether this joint is valid for visualization; skeleton: the pre-defined hand skeleton which can directly loaded from the given file.

If you have any other questions, feel free to ask me！