MotionWavelet: Human Motion Prediction via Wavelet Manifold Learning

TL;DR

MotionWavelet introduces a novel framework for human motion prediction that leverages wavelet transformations and diffusion models to better capture complex spatial-temporal motion patterns, resulting in improved accuracy and generalization.

Contribution

It proposes a Wavelet Diffusion Model and a Wavelet Space Shaping Guidance mechanism for more effective human motion prediction in the spatial-frequency domain.

Findings

Enhanced prediction accuracy across benchmarks

Improved generalization to diverse motion data

Effective modeling of non-stationary motion dynamics

Abstract

Modeling temporal characteristics and the non-stationary dynamics of body movement plays a significant role in predicting human future motions. However, it is challenging to capture these features due to the subtle transitions involved in the complex human motions. This paper introduces MotionWavelet, a human motion prediction framework that utilizes Wavelet Transformation and studies human motion patterns in the spatial-frequency domain. In MotionWavelet, a Wavelet Diffusion Model (WDM) learns a Wavelet Manifold by applying Wavelet Transformation on the motion data therefore encoding the intricate spatial and temporal motion patterns. Once the Wavelet Manifold is built, WDM trains a diffusion model to generate human motions from Wavelet latent vectors. In addition to the WDM, MotionWavelet also presents a Wavelet Space Shaping Guidance mechanism to refine the denoising process to improve conformity with the manifold structure. WDM also develops Temporal Attention-Based Guidance to enhance prediction accuracy. Extensive experiments validate the effectiveness of MotionWavelet, demonstrating improved prediction accuracy and enhanced generalization across various benchmarks. Our code and models will be released upon acceptance.