DiMo用离散扩散模型统一生成与理解动作，支持文本与无文本场景。
DiMo: Discrete Diffusion Modeling for Motion Generation and Understanding
- 采用迭代掩码重构机制，替代传统自回归解码。
- 在HumanML3D上生成动作质量优于基线，支持多任务统一建模。
- 适合需要灵活控制与高质量动作生成的研究者使用。
先前的掩码建模动作生成方法主要聚焦于文本到动作。本文提出DiMo，一种基于离散扩散风格的框架，将掩码建模扩展至双向文本-动作理解与生成。不同于GPT式自回归方法对动作进行逐帧编码，DiMo通过迭代掩码标记重构，在单个模型中统一实现文本到动作（T2M）、动作到文本（M2T）以及无文本动作到动作（M2M）。该解码范式可通过对重构步数调节实现质量与推理延迟的权衡。我们进一步利用残差向量量化（RVQ）提升动作标记保真度，并通过组相对策略优化（GRPO）增强对齐性与可控性。在HumanML3D和KIT-ML数据集上的实验表明，该框架在统一架构下实现了优异的动作质量与竞争力的双向理解能力。此外，无需修改结构即可完成无文本动作补全、文本引导动作预测及动作描述修正。更多定性结果见项目页：https://animotionlab.github.io/DiMo/。
原文摘要 · Abstract (English)
Prior masked modeling motion generation methods predominantly study text-to-motion. We present DiMo, a discrete diffusion-style framework, which extends masked modeling to bidirectional text--motion understanding and generation. Unlike GPT-style autoregressive approaches that tokenize motion and decode sequentially, DiMo performs iterative masked token refinement, unifying Text-to-Motion (T2M), Motion-to-Text (M2T), and text-free Motion-to-Motion (M2M) within a single model. This decoding paradigm naturally enables a quality-latency trade-off at inference via the number of refinement steps. We further improve motion token fidelity with residual vector quantization (RVQ) and enhance alignment and controllability with Group Relative Policy Optimization (GRPO). Experiments on HumanML3D and KIT-ML show strong motion quality and competitive bidirectional understanding under a unified framework. In addition, we demonstrate model ability in text-free motion completion, text-guided motion prediction and motion caption correction without architectural change. Additional qualitative results are available on our project page: https://animotionlab.github.io/DiMo/.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。