ContactOpt: Optimizing Contact to Improve Grasps
Abstract
Physical contact between hands and objects plays a critical role in human grasps. We show that optimizing the pose of a hand to achieve expected contact with an object can improve hand poses inferred via image-based methods. Given a hand mesh and an object mesh, a deep model trained on ground truth contact data infers desirable contact across the surfaces of the meshes. Then, ContactOpt efficiently optimizes the pose of the hand to achieve desirable contact using a differentiable contact model. Notably, our contact model encourages mesh interpenetration to approximate deformable soft tissue in the hand. In our evaluations, our methods result in grasps that better match ground truth contact, have lower kinematic error, and are significantly preferred by human participants. Code and models are available online.
Cite
Text
Grady et al. "ContactOpt: Optimizing Contact to Improve Grasps." Conference on Computer Vision and Pattern Recognition, 2021. doi:10.1109/CVPR46437.2021.00152Markdown
[Grady et al. "ContactOpt: Optimizing Contact to Improve Grasps." Conference on Computer Vision and Pattern Recognition, 2021.](https://mlanthology.org/cvpr/2021/grady2021cvpr-contactopt/) doi:10.1109/CVPR46437.2021.00152BibTeX
@inproceedings{grady2021cvpr-contactopt,
title = {{ContactOpt: Optimizing Contact to Improve Grasps}},
author = {Grady, Patrick and Tang, Chengcheng and Twigg, Christopher D. and Vo, Minh and Brahmbhatt, Samarth and Kemp, Charles C.},
booktitle = {Conference on Computer Vision and Pattern Recognition},
year = {2021},
pages = {1471-1481},
doi = {10.1109/CVPR46437.2021.00152},
url = {https://mlanthology.org/cvpr/2021/grady2021cvpr-contactopt/}
}