Year: 2025
Type: preprint
Abstract: This paper introduces VimoRAG, a novel video-based retrieval-augmented motion generation framework for motion large language models (LLMs). As motion LLMs face severe out-of-domain/out-of-vocabulary issues due to limited annotated data, VimoRAG leverages large-scale in-the-wild video databases to enhance 3D motion generation by retrieving relevant 2D human motion signals. While video-based motion RAG is nontrivial, we address two key bottlenecks: (1) developing an effective motion-centered video retrieval model that distinguishes human poses and actions, and (2) mitigating the issue of error propagation caused by suboptimal retrieval results. We design the Gemini Motion Video Retriever mechanism and the Motion-centric Dual-alignment DPO Trainer, enabling effective retrieval and generation processes. Experimental results show that VimoRAG significantly boosts the performance of motion LLMs constrained to text-only input. All the resources are available at https://walkermitty.github.io/VimoRAG/
Source: arXiv (Cornell University)
Authors: Xu, Haidong, Xu, Guangwei, Zheng, Zhedong, Zhu, Xiatian, Ji, Wei, Li, Xiangtai, Guo, Ruijie, Zhang, Meishan, zhang, Min, Fei, Hao
Language: English
Topic: Human Motion and Animation
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7076954109
Cites:
Cited by:
Related to: