InterEdit: Navigating Text-Guided Multi-Human 3D Motion Editing

TL;DR

This paper introduces InterEdit3D, a new dataset and benchmark for multi-person 3D motion editing guided by text, along with a novel diffusion model that captures interaction cues and motion dynamics, advancing multi-human motion editing capabilities.

Contribution

The paper presents a new dataset, benchmark, and a diffusion-based model for multi-person 3D motion editing guided by text, addressing the complexity of inter-person interactions.

Findings

InterEdit improves text-to-motion consistency.

InterEdit achieves state-of-the-art performance in multi-human motion editing.

The dataset and code are publicly released.

Abstract

Text-guided 3D motion editing has seen success in single-person scenarios, but its extension to multi-person settings is less explored due to limited paired data and the complexity of inter-person interactions. We introduce the task of multi-person 3D motion editing, where a target motion is generated from a source and a text instruction. To support this, we propose InterEdit3D, a new dataset with manual two-person motion change annotations, and a Text-guided Multi-human Motion Editing (TMME) benchmark. We present InterEdit, a synchronized classifier-free conditional diffusion model for TMME. It introduces Semantic-Aware Plan Token Alignment with learnable tokens to capture high-level interaction cues and an Interaction-Aware Frequency Token Alignment strategy using DCT and energy pooling to model periodic motion dynamics. Experiments show that InterEdit improves text-to-motion consistency and edit fidelity, achieving state-of-the-art TMME performance. The dataset and code will be released at https://github.com/YNG916/InterEdit.