ReMoMask：检索增强的掩码运动生成

摘要

文本到动作生成（Text-to-Motion, T2M）旨在从自然语言描述中合成出真实且语义对齐的人体运动序列。然而，现有方法面临双重挑战：生成模型（如扩散模型）受限于多样性不足、误差累积及物理不真实性，而检索增强生成（Retrieval-Augmented Generation, RAG）方法则表现出扩散惰性、部分模式崩溃及异步伪影。为克服这些局限，我们提出了ReMoMask，一个融合三项关键创新的统一框架：1）双向动量文本-动作模型通过动量队列解耦负样本规模与批量大小，显著提升跨模态检索精度；2）语义时空注意力机制在部分层级融合中强化生物力学约束，以消除异步伪影；3）RAG-无分类器指导结合少量无条件生成，增强泛化能力。基于MoMask的RVQ-VAE，ReMoMask在极少的步骤内高效生成时间连贯的动作。在标准基准上的大量实验验证了ReMoMask的顶尖性能，相较于前SOTA方法RAG-T2M，在HumanML3D和KIT-ML上的FID分数分别提升了3.88%和10.97%。代码与网站链接如下：https://github.com/AIGeeksGroup/ReMoMask，https://aigeeksgroup.github.io/ReMoMask。

English

Text-to-Motion (T2M) generation aims to synthesize realistic and semantically aligned human motion sequences from natural language descriptions. However, current approaches face dual challenges: Generative models (e.g., diffusion models) suffer from limited diversity, error accumulation, and physical implausibility, while Retrieval-Augmented Generation (RAG) methods exhibit diffusion inertia, partial-mode collapse, and asynchronous artifacts. To address these limitations, we propose ReMoMask, a unified framework integrating three key innovations: 1) A Bidirectional Momentum Text-Motion Model decouples negative sample scale from batch size via momentum queues, substantially improving cross-modal retrieval precision; 2) A Semantic Spatio-temporal Attention mechanism enforces biomechanical constraints during part-level fusion to eliminate asynchronous artifacts; 3) RAG-Classier-Free Guidance incorporates minor unconditional generation to enhance generalization. Built upon MoMask's RVQ-VAE, ReMoMask efficiently generates temporally coherent motions in minimal steps. Extensive experiments on standard benchmarks demonstrate the state-of-the-art performance of ReMoMask, achieving a 3.88% and 10.97% improvement in FID scores on HumanML3D and KIT-ML, respectively, compared to the previous SOTA method RAG-T2M. Code: https://github.com/AIGeeksGroup/ReMoMask. Website: https://aigeeksgroup.github.io/ReMoMask.

ReMoMask：检索增强的掩码运动生成

ReMoMask: Retrieval-Augmented Masked Motion Generation

摘要

Support