MoRAG -- Multi-Fusion Retrieval Augmented Generation for Human Motion

TL;DR

MoRAG introduces a multi-fusion retrieval-augmented generation approach that enhances human motion generation by improving retrieval accuracy, diversity, and generalization through multi-part strategies and large language model prompting.

Contribution

The paper presents a novel multi-part retrieval strategy combined with LLM prompting to improve motion retrieval and generation, especially for unseen text descriptions.

Findings

Enhanced motion generation performance demonstrated in experiments.

Effective handling of spelling errors and rephrasing in retrieval.

Improved generalizability across language space.

Abstract

We introduce MoRAG, a novel multi-part fusion based retrieval-augmented generation strategy for text-based human motion generation. The method enhances motion diffusion models by leveraging additional knowledge obtained through an improved motion retrieval process. By effectively prompting large language models (LLMs), we address spelling errors and rephrasing issues in motion retrieval. Our approach utilizes a multi-part retrieval strategy to improve the generalizability of motion retrieval across the language space. We create diverse samples through the spatial composition of the retrieved motions. Furthermore, by utilizing low-level, part-specific motion information, we can construct motion samples for unseen text descriptions. Our experiments demonstrate that our framework can serve as a plug-and-play module, improving the performance of motion diffusion models. Code, pretrained models and sample videos are available at: https://motion-rag.github.io/