Data-Driven Approach to Simulating Realistic Human Joint Constraints

TL;DR

This paper presents a neural network-based method to learn pose-dependent human joint limits from real data, enabling more realistic physics simulation of human joints in robotics and animation.

Contribution

It introduces an implicit neural function to accurately model human joint constraints, improving simulation realism over traditional fixed-limit approaches.

Findings

Successfully learned pose-dependent joint limits from real human data.

Efficiently enforced joint constraints using gradient-based methods.

Enhanced realism in physics simulations of human movement.

Abstract

Modeling realistic human joint limits is important for applications involving physical human-robot interaction. However, setting appropriate human joint limits is challenging because it is pose-dependent: the range of joint motion varies depending on the positions of other bones. The paper introduces a new technique to accurately simulate human joint limits in physics simulation. We propose to learn an implicit equation to represent the boundary of valid human joint configurations from real human data. The function in the implicit equation is represented by a fully connected neural network whose gradients can be efficiently computed via back-propagation. Using gradients, we can efficiently enforce realistic human joint limits through constraint forces in a physics engine or as constraints in an optimization problem.