MambaGesture: Enhancing Co-Speech Gesture Generation with Mamba and Disentangled Multi-Modality Fusion

TL;DR

MambaGesture introduces a novel multi-modality fusion framework with Mamba-based attention to generate more realistic, diverse, and synchronized co-speech gestures, advancing avatar animation quality.

Contribution

The paper proposes MambaGesture, combining Mamba attention and disentangled multi-modality fusion, to improve gesture realism and diversity in co-speech gesture generation.

Findings

Achieved state-of-the-art Fréchet Gesture Distance (FGD) scores.

Enhanced gesture diversity and beat alignment accuracy.

Demonstrated robustness across the BEAT dataset.

Abstract

Co-speech gesture generation is crucial for producing synchronized and realistic human gestures that accompany speech, enhancing the animation of lifelike avatars in virtual environments. While diffusion models have shown impressive capabilities, current approaches often overlook a wide range of modalities and their interactions, resulting in less dynamic and contextually varied gestures. To address these challenges, we present MambaGesture, a novel framework integrating a Mamba-based attention block, MambaAttn, with a multi-modality feature fusion module, SEAD. The MambaAttn block combines the sequential data processing strengths of the Mamba model with the contextual richness of attention mechanisms, enhancing the temporal coherence of generated gestures. SEAD adeptly fuses audio, text, style, and emotion modalities, employing disentanglement to deepen the fusion process and yield gestures with greater realism and diversity. Our approach, rigorously evaluated on the multi-modal BEAT dataset, demonstrates significant improvements in Fr\'echet Gesture Distance (FGD), diversity scores, and beat alignment, achieving state-of-the-art performance in co-speech gesture generation. Project website: $\href{https://fcchit.github.io/mambagesture/}{\textit{https://fcchit.github.io/mambagesture/}}$.