BoDiffusion: Diffusing Sparse Observations for Full-Body Human Motion Synthesis

TL;DR

BoDiffusion is a novel generative diffusion model that reconstructs full-body human motion from limited head and hand tracking data, enabling more immersive mixed reality experiences.

Contribution

It introduces the first reverse diffusion-based approach for full-body motion reconstruction from sparse observations, improving realism and accuracy.

Findings

Outperforms state-of-the-art methods in motion realism

Reduces joint reconstruction error significantly

Effective in large-scale motion-capture datasets

Abstract

Mixed reality applications require tracking the user's full-body motion to enable an immersive experience. However, typical head-mounted devices can only track head and hand movements, leading to a limited reconstruction of full-body motion due to variability in lower body configurations. We propose BoDiffusion -- a generative diffusion model for motion synthesis to tackle this under-constrained reconstruction problem. We present a time and space conditioning scheme that allows BoDiffusion to leverage sparse tracking inputs while generating smooth and realistic full-body motion sequences. To the best of our knowledge, this is the first approach that uses the reverse diffusion process to model full-body tracking as a conditional sequence generation task. We conduct experiments on the large-scale motion-capture dataset AMASS and show that our approach outperforms the state-of-the-art approaches by a significant margin in terms of full-body motion realism and joint reconstruction error.