Paper 2512.18181
MACE-Dance: Motion-Appearance Cascaded Experts for Music-Driven Dance Video Generation
- Published
- Dec 2025
- Research lab
- Independent
- Citations
- 9
- GitHub
- 108 stars
01 In brief
Summary
MACE-Dance is a music-driven dance video generation framework using a cascaded Mixture-of-Experts (MoE) design, decoupling the task into a Motion Expert and an Appearance Expert.
The Motion Expert generates 3D SMPL motion from music using a Diffusion Model with a BiMamba-Transformer hybrid architecture and Guidance-Free Training (GFT), achieving state-of-the-art (SOTA) performance on the FineDance dataset.
The Appearance Expert, built on Wan-Animate, uses a decoupled Kinematic–Aesthetic fine-tuning strategy to synthesize videos from the 3D motion and a reference image, achieving SOTA on the MA-Data dataset for pose-driven image animation.
The authors curated MA-Data, a large-scale dataset of 70k clips (116 hours) spanning 20+ dance genres, and designed a motion–appearance evaluation protocol.
Experiments show MACE-Dance achieves SOTA on music-driven dance video generation, outperforming baselines like Hallo2, WAN-S2V, and MEGA in both motion quality (e.g., FID_k=16.46, BAS=0.523) and appearance quality (e.g., SC=93.97, TF=97.10).
The framework also supports long-sequence generation and motion editing.
Future work includes text-controlled generation and efficiency improvements.
02 From the paper
Abstract
With the rise of online dance-video platforms and rapid advances in AI-generated content (AIGC), music-driven dance generation has emerged as a compelling research direction. Despite substantial progress in related domains such as music-driven 3D dance generation, pose-driven image animation, and audio-driven talking-head synthesis, existing methods cannot be directly adapted to this task. Moreover, the limited studies in this area still struggle to jointly achieve high-quality visual appearance and realistic human motion. Accordingly, we present MACE-Dance, a music-driven dance video generation framework with cascaded Mixture-of-Experts (MoE). The Motion Expert performs music-to-3D motion generation while enforcing kinematic plausibility and artistic expressiveness, whereas the Appearance Expert carries out motion- and reference-conditioned video synthesis, preserving visual identity with spatiotemporal coherence. Specifically, the Motion Expert adopts a diffusion model with a BiMamba-Transformer hybrid architecture and a Guidance-Free Training (GFT) strategy, achieving state-of-the-art (SOTA) performance in 3D dance generation. The Appearance Expert employs a decoupled kinematic-aesthetic fine-tuning strategy, achieving state-of-the-art (SOTA) performance in pose-driven image animation. To better benchmark this task, we curate a large-scale and diverse dataset and design a motion-appearance evaluation protocol. Based on this protocol, MACE-Dance also achieves state-of-the-art performance. Code is available at https://github.com/AMAP-ML/MACE-Dance.