Downloads · 30 days
0
qq456cvb/UKPGAN
UKPGAN is a keypoint detection model from qq456cvb. Use it for the keypoint detection task on the model card, and read the license before you ship it in a product. It is set up for tensorflow. The card lists the license as mit.
Pretrained TensorFlow checkpoints for UKPGAN, a self-supervised 3D keypoint detector on rigid/non-rigid objects and real scenes, from:
Downloads · 30 days
0
Access
Public
Updated Oct 6, 2026
Repo size
714 MB
Likes
0
Public
Click a slice to open those files.
.data-00000-of-00001714 MB · 100%
From the Hugging Face model README
Pretrained TensorFlow checkpoints for UKPGAN, a self-supervised 3D keypoint detector on rigid/non-rigid objects and real scenes, from:
UKPGAN: A General Self-Supervised Keypoint Detector (CVPR 2022)
Paper: arXiv:2011.11974 · Hugging Face paper page
Code and usage instructions: https://github.com/qq456cvb/UKPGAN
Project page: https://qq456cvb.github.io/projects/ukpgan
Checkpoints under outputs/<category>/tflogs/, matching the layout expected by the code:
airplane, chair, table — ShapeNet category models.smpl — non-rigid SMPL human models.universal — trained on a large collection of ShapeNet models (10+ categories), used for real-world scenes (3DMatch).Download into the root of the UKPGAN repo:
hf download qq456cvb/UKPGAN --local-dir .
python visualize.py --type shapenet --nms --nms_radius 0.1
@inproceedings{you2022ukpgan,
title={UKPGAN: A General Self-Supervised Keypoint Detector},
author={You, Yang and Liu, Wenhai and Ze, Yanjie and Li, Yong-Lu and Wang, Weiming and Lu, Cewu},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year={2022}
}