Motion Mamba: Efficient and Long Sequence Motion Generation

TL;DR

Motion Mamba introduces a novel hierarchical and bidirectional SSM-based model for efficient, high-quality long-sequence human motion generation, significantly outperforming previous diffusion methods in speed and accuracy.

Contribution

It pioneers the use of SSMs in motion generation with a hierarchical U-Net architecture and bidirectional pose processing, addressing previous limitations.

Findings

Achieves up to 50% FID improvement over previous methods.

Runs up to 4 times faster than diffusion-based models.

Demonstrates strong long-sequence motion modeling capabilities.

Abstract

Human motion generation stands as a significant pursuit in generative computer vision, while achieving long-sequence and efficient motion generation remains challenging. Recent advancements in state space models (SSMs), notably Mamba, have showcased considerable promise in long sequence modeling with an efficient hardware-aware design, which appears to be a promising direction to build motion generation model upon it. Nevertheless, adapting SSMs to motion generation faces hurdles since the lack of a specialized design architecture to model motion sequence. To address these challenges, we propose Motion Mamba, a simple and efficient approach that presents the pioneering motion generation model utilized SSMs. Specifically, we design a Hierarchical Temporal Mamba (HTM) block to process temporal data by ensemble varying numbers of isolated SSM modules across a symmetric U-Net architecture aimed at preserving motion consistency between frames. We also design a Bidirectional Spatial Mamba (BSM) block to bidirectionally process latent poses, to enhance accurate motion generation within a temporal frame. Our proposed method achieves up to 50% FID improvement and up to 4 times faster on the HumanML3D and KIT-ML datasets compared to the previous best diffusion-based method, which demonstrates strong capabilities of high-quality long sequence motion modeling and real-time human motion generation. See project website https://steve-zeyu-zhang.github.io/MotionMamba/