Variational Autoencoder with Differentiable Physics Engine for Human Gait Analysis and Synthesis
Abstract
We address the task of learning generative models of human gait. As gait motion always follows the physical laws, a generative model should also produce outputs that comply with the physical laws, particularly rigid body dynamics with contact and friction. We propose a deep generative model combined with a differentiable physics engine, which outputs physically plausible signals by construction. The proposed model is also equipped with a policy network conditioned on each sample. We show an example of the application of such a model to style transfer of gait.
Cite
Text
Takeishi and Kalousis. "Variational Autoencoder with Differentiable Physics Engine for Human Gait Analysis and Synthesis." NeurIPS 2021 Workshops: DGMs_Applications, 2021.Markdown
[Takeishi and Kalousis. "Variational Autoencoder with Differentiable Physics Engine for Human Gait Analysis and Synthesis." NeurIPS 2021 Workshops: DGMs_Applications, 2021.](https://mlanthology.org/neuripsw/2021/takeishi2021neuripsw-variational/)BibTeX
@inproceedings{takeishi2021neuripsw-variational,
title = {{Variational Autoencoder with Differentiable Physics Engine for Human Gait Analysis and Synthesis}},
author = {Takeishi, Naoya and Kalousis, Alexandros},
booktitle = {NeurIPS 2021 Workshops: DGMs_Applications},
year = {2021},
url = {https://mlanthology.org/neuripsw/2021/takeishi2021neuripsw-variational/}
}