RMD: A Simple Baseline for More General Human Motion Generation via Training-free Retrieval-Augmented Motion Diffuse
This addresses the challenge of handling out-of-distribution scenarios in motion generation for practical applications, though it is incremental as it builds on existing retrieval and diffusion methods.
The paper tackles the problem of limited dataset diversity and scale in human motion generation by proposing RMD, a training-free retrieval-augmented baseline that enhances generalization, achieving state-of-the-art performance with notable advantages on out-of-distribution data.
While motion generation has made substantial progress, its practical application remains constrained by dataset diversity and scale, limiting its ability to handle out-of-distribution scenarios. To address this, we propose a simple and effective baseline, RMD, which enhances the generalization of motion generation through retrieval-augmented techniques. Unlike previous retrieval-based methods, RMD requires no additional training and offers three key advantages: (1) the external retrieval database can be flexibly replaced; (2) body parts from the motion database can be reused, with an LLM facilitating splitting and recombination; and (3) a pre-trained motion diffusion model serves as a prior to improve the quality of motions obtained through retrieval and direct combination. Without any training, RMD achieves state-of-the-art performance, with notable advantages on out-of-distribution data.