MotionHiFlow：基于层次化流匹配的文本驱动运动生成
MotionHiFlow: Text-to-Motion via Hierarchical Flow Matching
把流匹配的采样路径按时间尺度分层（低尺度定语义骨架、高尺度补细节），并用跨尺度过渡保持噪声一致性，这一「由粗到细的层次化流路径」思路可直接迁移到视频生成、长时序动作合成等需要兼顾语义对齐与细粒度时序一致性的视觉生成任务。
原文依据 a hierarchical flow matching framework to generate motion progressively by constructing flow path from low to high temporal scales
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论文本驱动运动生成旨在生成与输入文本紧密对齐、同时保持物理合理性并富含细粒度细节的三维人体运动。尽管近期方法能够产生复杂而自然的动作，但它们通常只在单一时间尺度上运作，这同时限制了语义对齐与时序连贯性。受人类认知系统中复杂运动是在多个时间尺度上分层概念化、而非单一时间尺度这一事实的启发，我们提出 MotionHiFlow，一个层次化流匹配框架，通过构造从低时间尺度到高时间尺度的流路径来渐进式生成运动。较低尺度上的流捕捉高层语义与粗粒度运动结构，而较高尺度上的流则细化时序细节。为连接跨尺度的流，我们引入一种新颖的跨尺度过渡过程，确保连续性并保持噪声一致性。此外，通过整合文本-运动扩散 Transformer 与拓扑感知的运动 VAE，MotionHiFlow 借助关节感知位置编码与骨架拓扑显式建模关节间的结构依赖关系，从而在实现精确语义对齐的同时保留细粒度运动细节。在 HumanML3D 与 KIT-ML 基准上的大量实验展示了最优性能，消融研究证实了层次化设计及关键组件的有效性。
英文原文摘要 1432 字符
Text-to-motion generation aims to generate 3D human motions that are tightly aligned with the input text while remaining physically plausible and rich in fine-grained detail. Although recent approaches can produce complex and natural movements, they usually operate at only one temporal scale, which limits both semantic alignment and temporal coherence. Inspired by the fact that complex motions are conceptualized hierarchically rather than at a single temporal scale in the human cognitive system, we propose MotionHiFlow, a hierarchical flow matching framework to generate motion progressively by constructing flow path from low to high temporal scales. The flows at lower scales capture high-level semantics and coarse motion structures, while flows at higher scales refine temporal details. To link the flows across scales, we introduce a novel cross-scale transition process, ensuring continuity and preserving noise consistency. Furthermore, by integrating a Text-Motion Diffusion Transformer and a topology-aware Motion VAE, MotionHiFlow explicitly models structural dependencies among joints via joint-aware positional encoding and skeletal topology, enabling precise semantic alignment alongside fine-grained motion details. Extensive experiments on HumanML3D and KIT-ML benchmarks demonstrate state-of-the-art performance, with ablation studies confirming the effectiveness of the hierarchical design and key components.