Variational Autoencoder with Differentiable Physics Engine for Human Gait Analysis and Synthesis

Naoya Takeishi, Alexandros Kalousis

Keywords: variational autoencoder, differentiable physics engine, human gait

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.

TL;DR: We use a generative model integrated with a differentiable physics engine for modeling human gait.