NASA Neural Articulated Shape Approximation

Abstract

Efficient representation of articulated objects such as human bodies is an important problem in computer vision and graphics. To efficiently simulate deformation, existing approaches represent 3D objects using polygonal meshes and deform them using skinning techniques. This paper introduces neural articulated shape approximation (NASA), an alternative framework that enables efficient representation of articulated deformable objects using neural indicator functions that are conditioned on pose. Occupancy testing using NASA is straightforward, circumventing the complexity of meshes and the issue of water-tightness. We demonstrate the effectiveness of NASA for 3D tracking applications, and discuss other potential extensions.

Cite

Text

Deng et al. "NASA Neural Articulated Shape Approximation." Proceedings of the European Conference on Computer Vision (ECCV), 2020. doi:10.1007/978-3-030-58571-6_36

Markdown

[Deng et al. "NASA Neural Articulated Shape Approximation." Proceedings of the European Conference on Computer Vision (ECCV), 2020.](https://mlanthology.org/eccv/2020/deng2020eccv-nasa/) doi:10.1007/978-3-030-58571-6_36

BibTeX

@inproceedings{deng2020eccv-nasa,
  title     = {{NASA Neural Articulated Shape Approximation}},
  author    = {Deng, Boyang and Lewis, Jp and Jeruzalski, Timothy and Pons-Moll, Gerard and Hinton, Geoffrey and Norouzi, Mohammad and Tagliasacchi, Andrea},
  booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
  year      = {2020},
  doi       = {10.1007/978-3-030-58571-6_36},
  url       = {https://mlanthology.org/eccv/2020/deng2020eccv-nasa/}
}