2025/05/14 by Julian Tanke, Tanke, Julian, Takashi Shibuya +7 · 3 citations
Psychology · #Benchmark (surveying) #Computer Vision and Pattern Recognition (cs.CV) #Encoding (memory) #Ergonomics and Musculoskeletal Disorders #FOS: Computer and information sciences #Human motion #Motion (physics) #Sequence (biology) #Task (project management)
paper · pdf · doi:10.48550/arxiv.2505.09827
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2025/05/14 · openalex created_date 2025/10/15 · openalex updated_date 2026/08/05
Generating realistic dyadic human motion from text descriptions presents significant challenges, particularly for extended interactions that exceed typical training sequence lengths. While recent transformer-based approaches have shown promising results for short-term dyadic motion synthesis, they struggle with longer sequences due to inherent limitations in positional encoding schemes. In this paper, we introduce Dyadic Mamba, a novel approach that leverages State-Space Models (SSMs) to generate high-quality dyadic human motion of arbitrary length. Our method employs a simple yet effective architecture that facilitates information flow between individual motion sequences through concatenation, eliminating the need for complex cross-attention mechanisms. We demonstrate that Dyadic Mamba achieves competitive performance on standard short-term benchmarks while significantly outperforming transformer-based approaches on longer sequences. Additionally, we propose a new benchmark for evaluating long-term motion synthesis quality, providing a standardized framework for future research. Our results demonstrate that SSM-based architectures offer a promising direction for addressing the challenging task of long-term dyadic human motion synthesis from text descriptions.