FlowerDance: MeanFlow for Efficient and Refined 3D Dance Generation

TL;DR

FlowerDance introduces an efficient, high-quality 3D dance generation method that combines MeanFlow with physical constraints, enabling fast inference and interactive editing for realistic dance synthesis.

Contribution

It proposes FlowerDance, a novel framework that enhances 3D dance generation efficiency and quality using MeanFlow, physical constraints, and a simple architecture with motion editing capabilities.

Findings

Achieves state-of-the-art motion quality on AIST++ and FineDance datasets.

Significantly improves inference speed and memory efficiency.

Supports interactive motion editing for refined dance sequences.

Abstract

Music-to-dance generation aims to translate auditory signals into expressive human motion, with broad applications in virtual reality, choreography, and digital entertainment. Despite promising progress, the limited generation efficiency of existing methods leaves insufficient computational headroom for high-fidelity 3D rendering, thereby constraining the expressiveness of 3D characters during real-world applications. Thus, we propose FlowerDance, which not only generates refined motion with physical plausibility and artistic expressiveness, but also achieves significant generation efficiency on inference speed and memory utilization. Specifically, FlowerDance combines MeanFlow with Physical Consistency Constraints, which enables high-quality motion generation with only a few sampling steps. Moreover, FlowerDance leverages a simple but efficient model architecture with BiMamba-based backbone and Channel-Level Cross-Modal Fusion, which generates dance with efficient non-autoregressive manner. Meanwhile, FlowerDance supports motion editing, enabling users to interactively refine dance sequences. Extensive experiments on AIST++ and FineDance show that FlowerDance achieves state-of-the-art results in both motion quality and generation efficiency. Code will be released upon acceptance. Project page: https://flowerdance25.github.io/ .