Dance Dataset
Motion tokens, motion features, SMPL-X parameters, music features, and evaluation data.
Browse dataset ↗Dataset & Resources
Get the released dance data and model weights, or generate a dance from your own music.
Pretrained Models
ChoreoLLaMA generation weights, with the VQ-VAE and RetrievalNet checkpoints available in the same release.
Browse model weights ↗Run InfiniteDance
Follow the setup and download guide, then run inference on the test set or your own audio.
Get started ↗Released for non-commercial academic research, education, and evaluation. See the license and third-party notices for usage terms. The download guide lists file sizes and extraction paths.
Abstract
Although existing 3D dance generation methods perform well in controlled scenarios, they often struggle to generalize in the wild. When conditioned on unseen music, existing methods often produce unstructured or physically implausible dance, largely due to limited music-to-dance data and restricted model capacity. This work aims to push the frontier of generalizable 3D dance generation by scaling up both data and model design. 1) On the data side, we develop a fully automated pipeline that reconstructs high-fidelity 3D dance motions from monocular videos. To eliminate the physical artifacts prevalent in existing reconstruction methods, we introduce a Foot Restoration Diffusion Model (FRDM) guided by foot-contact and geometric constraints that enforce physical plausibility while preserving kinematic smoothness and expressiveness, resulting in a diverse, high-quality multimodal 3D dance dataset. 2) On model design, we propose Choreographic LLaMA (ChoreoLLaMA), a scalable LLaMA-based architecture. To enhance robustness under unfamiliar music conditions, we integrate a retrieval-augmented generation (RAG) module that injects reference dance as a prompt. Additionally, we design a slow/fast-cadence Mixture-of-Experts (MoE) module that enables ChoreoLLaMA to smoothly adapt motion rhythms across varying music tempos. Extensive experiments across diverse dance genres show that our approach surpasses existing methods in both qualitative and quantitative evaluations, marking a step toward scalable, real-world 3D dance generation.
Method
In order to achieve scalable and generalizable 3D dance generation, we design a two-pronged approach that addresses both data quality and model architecture.
Data Pipeline with Foot Restoration
We develop a fully automated pipeline that reconstructs high-fidelity 3D dance motions from monocular videos. To eliminate physical artifacts prevalent in existing reconstruction methods, we introduce a Foot Restoration Diffusion Model (FRDM) guided by foot-contact and geometric constraints. This ensures physical plausibility while preserving kinematic smoothness and expressiveness, resulting in a diverse, high-quality multimodal 3D dance dataset.
Choreographic LLaMA Architecture
We propose Choreographic LLaMA (ChoreoLLaMA), a scalable LLaMA-based architecture for 3D dance generation. To enhance robustness under unfamiliar music conditions, we integrate a retrieval-augmented generation (RAG) module that injects reference dance as a prompt. Additionally, we design a slow/fast-cadence Mixture-of-Experts (MoE) module that enables ChoreoLLaMA to smoothly adapt motion rhythms across varying music tempos.
Results
Watch ChoreoLLaMA generate 3D dance from music across a range of styles. Press play and unmute to hear the accompanying music.
Citation
If you use InfiniteDance in your research, please cite our work.
@misc{li2026infinitedancescalable3ddance,
title={InfiniteDance: Scalable 3D Dance Generation Towards in-the-wild Generalization},
author={Ronghui Li and Zhongyuan Hu and Li Siyao and Youliang Zhang and Haozhe Xie and Mingyuan Zhang and Jie Guo and Xiu Li and Ziwei Liu},
year={2026},
eprint={2603.13375},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2603.13375},
}Download BibTeX · ECCV 2026 conference page ↗
For the integrated motion corpus, please also cite the source datasets listed in the repository citation guide.