MotionFlux：通过校正流匹配与偏好对齐实现高效的文本引导运动生成

摘要

动作生成对于虚拟角色和具身代理的动画制作至关重要。尽管近期基于文本驱动的方法取得了显著进展，但它们往往难以实现语言描述与动作语义之间的精确对齐，同时也受限于缓慢、多步推理的低效性。为解决这些问题，我们引入了TMR++对齐偏好优化（TAPO），这是一个创新框架，能够将细微的动作变化与文本修饰符对齐，并通过迭代调整强化语义基础。为进一步实现实时合成，我们提出了MotionFLUX，一个基于确定性修正流匹配的高速生成框架。与需要数百步去噪的传统扩散模型不同，MotionFLUX在噪声分布与动作空间之间构建最优传输路径，从而促进实时合成。线性化的概率路径减少了对序列方法中多步采样的需求，在不牺牲动作质量的前提下显著加速了推理时间。实验结果表明，TAPO与MotionFLUX共同构成了一个统一系统，在语义一致性和动作质量上均超越了现有最先进方法，同时大幅提升了生成速度。代码及预训练模型将予以发布。

English

Motion generation is essential for animating virtual characters and embodied agents. While recent text-driven methods have made significant strides, they often struggle with achieving precise alignment between linguistic descriptions and motion semantics, as well as with the inefficiencies of slow, multi-step inference. To address these issues, we introduce TMR++ Aligned Preference Optimization (TAPO), an innovative framework that aligns subtle motion variations with textual modifiers and incorporates iterative adjustments to reinforce semantic grounding. To further enable real-time synthesis, we propose MotionFLUX, a high-speed generation framework based on deterministic rectified flow matching. Unlike traditional diffusion models, which require hundreds of denoising steps, MotionFLUX constructs optimal transport paths between noise distributions and motion spaces, facilitating real-time synthesis. The linearized probability paths reduce the need for multi-step sampling typical of sequential methods, significantly accelerating inference time without sacrificing motion quality. Experimental results demonstrate that, together, TAPO and MotionFLUX form a unified system that outperforms state-of-the-art approaches in both semantic consistency and motion quality, while also accelerating generation speed. The code and pretrained models will be released.

MotionFlux：通过校正流匹配与偏好对齐实现高效的文本引导运动生成

MotionFlux: Efficient Text-Guided Motion Generation through Rectified Flow Matching and Preference Alignment

摘要

Support