Dynamic Motion Blending for Versatile Motion Editing

2025-03-26

AI Summary

Existing text-guided motion editing methods suffer from poor generalization due to reliance on pre-collected motion-text triplets. This work proposes MotionCutMix, an online data augmentation strategy, and MotionReFit, an autoregressive diffusion model, introducing the first motion coordinator mechanism to resolve limb incoherence during dynamic motion splicing. By integrating online motion segment mixing, autoregressive temporal modeling, the motion coordinator module, and text-motion alignment representation learning, our framework enables high-degree-of-freedom spatiotemporal editing—without requiring keyframes, additional annotations, or large language models. Extensive experiments on multiple text-guided motion editing benchmarks demonstrate state-of-the-art performance, with significant improvements in editing diversity, physical plausibility, and semantic fidelity.

Technology Category

• Computer Vision: Motion & Tracking
• Machine Learning: Mixture of Experts (MoE)
• Natural Language Processing: Generation

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Text-guided motion editing enables high-level semantic control and iterative modifications beyond traditional keyframe animation. Existing methods rely on limited pre-collected training triplets, which severely hinders their versatility in diverse editing scenarios. We introduce MotionCutMix, an online data augmentation technique that dynamically generates training triplets by blending body part motions based on input text. While MotionCutMix effectively expands the training distribution, the compositional nature introduces increased randomness and potential body part incoordination. To model such a rich distribution, we present MotionReFit, an auto-regressive diffusion model with a motion coordinator. The auto-regressive architecture facilitates learning by decomposing long sequences, while the motion coordinator mitigates the artifacts of motion composition. Our method handles both spatial and temporal motion edits directly from high-level human instructions, without relying on additional specifications or Large Language Models. Through extensive experiments, we show that MotionReFit achieves state-of-the-art performance in text-guided motion editing.

Problem

• Expands motion editing versatility beyond limited training triplets
• Addresses incoordination in dynamically blended body part motions
• Enables text-guided spatial and temporal edits without LLMs

Innovation

• Online data augmentation with MotionCutMix
• Auto-regressive diffusion model MotionReFit
• Motion coordinator mitigates composition artifacts