ReactDance: Hierarchical Representation for High-Fidelity and Coherent Long-Form Reactive Dance Generation

TL;DR

ReactDance introduces a hierarchical diffusion framework with novel multi-scale motion representation and blockwise sampling to generate high-fidelity, coherent long-form reactive dance conditioned on a lead dancer's motion.

Contribution

The paper presents ReactDance, a hierarchical diffusion model with HFSQ and BLC strategies, enabling detailed spatial control and efficient long-term temporal coherence in reactive dance generation.

Findings

Outperforms state-of-the-art in motion quality

Achieves superior long-term coherence

Enhances sampling efficiency

Abstract

Reactive dance generation (RDG), the task of generating a dance conditioned on a lead dancer's motion, holds significant promise for enhancing human-robot interaction and immersive digital entertainment. Despite progress in duet synchronization and motion-music alignment, two key challenges remain: generating fine-grained spatial interactions and ensuring long-term temporal coherence. In this work, we introduce \textbf{ReactDance}, a diffusion framework that operates on a novel hierarchical latent space to address these spatiotemporal challenges in RDG. First, for high-fidelity spatial expression and fine-grained control, we propose Hierarchical Finite Scalar Quantization (\textbf{HFSQ}). This multi-scale motion representation effectively disentangles coarse body posture from subtle limb dynamics, enabling independent and detailed control over both aspects through a layered guidance mechanism. Second, to efficiently generate long sequences with high temporal coherence, we propose Blockwise Local Context (\textbf{BLC}), a non-autoregressive sampling strategy. Departing from slow, frame-by-frame generation, BLC partitions the sequence into blocks and synthesizes them in parallel via periodic causal masking and positional encodings. Coherence across these blocks is ensured by a dense sliding-window training approach that enriches the representation with local temporal context. Extensive experiments show that ReactDance substantially outperforms state-of-the-art methods in motion quality, long-term coherence, and sampling efficiency. Project page: https://ripemangobox.github.io/ReactDance.