From Universal Humanoid Control to Automatic Physically Valid Character Creation

TL;DR

This paper presents a framework for automatically designing physically valid humanoid bodies tailored to specific motions, leveraging a learned controller and physics simulation to optimize body shapes for desired movements.

Contribution

It introduces a novel design-and-control framework that combines a generalized humanoid controller with physics-based optimization to create custom humanoid bodies for specific motions.

Findings

Successfully generates humanoid designs for diverse motions

Outperforms baseline methods in motion imitation accuracy

Enables automatic customization of humanoid physical attributes

Abstract

Automatically designing virtual humans and humanoids holds great potential in aiding the character creation process in games, movies, and robots. In some cases, a character creator may wish to design a humanoid body customized for certain motions such as karate kicks and parkour jumps. In this work, we propose a humanoid design framework to automatically generate physically valid humanoid bodies conditioned on sequence(s) of pre-specified human motions. First, we learn a generalized humanoid controller trained on a large-scale human motion dataset that features diverse human motion and body shapes. Second, we use a design-and-control framework to optimize a humanoid's physical attributes to find body designs that can better imitate the pre-specified human motion sequence(s). Leveraging the pre-trained humanoid controller and physics simulation as guidance, our method is able to discover new humanoid designs that are customized to perform pre-specified human motions.