Physics-Guided Human Motion Capture with Pose Probability Modeling

TL;DR

This paper introduces a physics-guided approach for human motion capture that uses reverse diffusion and pose probability modeling to produce physically plausible motions, outperforming previous methods.

Contribution

The novel integration of physics as denoising guidance in reverse diffusion for monocular motion capture, addressing depth ambiguity and noise issues.

Findings

Outperforms previous physics-based methods in accuracy and success rate.

Uses a latent Gaussian model for 2D-to-3D pose uncertainty encoding.

Employs iterative physics-based tracking and kinematic denoising.

Abstract

Incorporating physics in human motion capture to avoid artifacts like floating, foot sliding, and ground penetration is a promising direction. Existing solutions always adopt kinematic results as reference motions, and the physics is treated as a post-processing module. However, due to the depth ambiguity, monocular motion capture inevitably suffers from noises, and the noisy reference often leads to failure for physics-based tracking. To address the obstacles, our key-idea is to employ physics as denoising guidance in the reverse diffusion process to reconstruct physically plausible human motion from a modeled pose probability distribution. Specifically, we first train a latent gaussian model that encodes the uncertainty of 2D-to-3D lifting to facilitate reverse diffusion. Then, a physics module is constructed to track the motion sampled from the distribution. The discrepancies between the tracked motion and image observation are used to provide explicit guidance for the reverse diffusion model to refine the motion. With several iterations, the physics-based tracking and kinematic denoising promote each other to generate a physically plausible human motion. Experimental results show that our method outperforms previous physics-based methods in both joint accuracy and success rate. More information can be found at \url{https://github.com/Me-Ditto/Physics-Guided-Mocap}.