English

Robust Dancer: Long-term 3D Dance Synthesis Using Unpaired Data

Computer Vision and Pattern Recognition 2023-03-30 v1 Graphics

Abstract

How to automatically synthesize natural-looking dance movements based on a piece of music is an incrementally popular yet challenging task. Most existing data-driven approaches require hard-to-get paired training data and fail to generate long sequences of motion due to error accumulation of autoregressive structure. We present a novel 3D dance synthesis system that only needs unpaired data for training and could generate realistic long-term motions at the same time. For the unpaired data training, we explore the disentanglement of beat and style, and propose a Transformer-based model free of reliance upon paired data. For the synthesis of long-term motions, we devise a new long-history attention strategy. It first queries the long-history embedding through an attention computation and then explicitly fuses this embedding into the generation pipeline via multimodal adaptation gate (MAG). Objective and subjective evaluations show that our results are comparable to strong baseline methods, despite not requiring paired training data, and are robust when inferring long-term music. To our best knowledge, we are the first to achieve unpaired data training - an ability that enables to alleviate data limitations effectively. Our code is released on https://github.com/BFeng14/RobustDancer

Keywords

Cite

@article{arxiv.2303.16856,
  title  = {Robust Dancer: Long-term 3D Dance Synthesis Using Unpaired Data},
  author = {Bin Feng and Tenglong Ao and Zequn Liu and Wei Ju and Libin Liu and Ming Zhang},
  journal= {arXiv preprint arXiv:2303.16856},
  year   = {2023}
}

Comments

Preliminary video demo: https://youtu.be/gJbxG9QlcUU

R2 v1 2026-06-28T09:40:21.773Z