InterDance: Reactive 3D Dance Generation with Realistic Duet Interactions

Humans can perform a variety of interactive motions, among which two-person dance is one of the most challenging interactions. However, in terms of computer motion generation, current work is still unable to generate high-quality interactive motion, especially in the field of duet dance. On the one hand, this is caused by the lack of large-scale high-quality datasets. On the other hand, it arises from the incomplete representation of interactive motion and the lack of fine-grained optimization of interactions. To address these challenges, we propose a duet dance dataset that significantly enhances motion quality, data scale, and the variety of dance genres. Based on this dataset, we propose a new motion representation that can accurately and comprehensively describe interactive motion. We further introduce a diffusion-based algorithm with an interaction refine guidance strategy to optimize the realism of interactions progressively. Experiments demonstrate the effectiveness of our dataset and algorithm.