Walk Before You Dance: High-fidelity and Editable Dance Synthesis via Generative Masked Motion Prior

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Shah, Foram N, Shah, Parshwa, Saleem, Muhammad Usama, Pinyoanuntapong, Ekkasit, Wang, Pu, Xue, Hongfei, Helmy, Ahmed
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915647490883584
author Shah, Foram N
Shah, Parshwa
Saleem, Muhammad Usama
Pinyoanuntapong, Ekkasit
Wang, Pu
Xue, Hongfei
Helmy, Ahmed
author_facet Shah, Foram N
Shah, Parshwa
Saleem, Muhammad Usama
Pinyoanuntapong, Ekkasit
Wang, Pu
Xue, Hongfei
Helmy, Ahmed
contents Recent advances in dance generation have enabled the automatic synthesis of 3D dance motions. However, existing methods still face significant challenges in simultaneously achieving high realism, precise dance-music synchronization, diverse motion expression, and physical plausibility. To address these limitations, we propose a novel approach that leverages a generative masked text-to-motion model as a distribution prior to learn a probabilistic mapping from diverse guidance signals, including music, genre, and pose, into high-quality dance motion sequences. Our framework also supports semantic motion editing, such as motion inpainting and body part modification. Specifically, we introduce a multi-tower masked motion model that integrates a text-conditioned masked motion backbone with two parallel, modality-specific branches: a music-guidance tower and a pose-guidance tower. The model is trained using synchronized and progressive masked training, which allows effective infusion of the pretrained text-to-motion prior into the dance synthesis process while enabling each guidance branch to optimize independently through its own loss function, mitigating gradient interference. During inference, we introduce classifier-free logits guidance and pose-guided token optimization to strengthen the influence of music, genre, and pose signals. Extensive experiments demonstrate that our method sets a new state of the art in dance generation, significantly advancing both the quality and editability over existing approaches. Project Page available at https://foram-s1.github.io/DanceMosaic/
format Preprint
id arxiv_https___arxiv_org_abs_2504_04634
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Walk Before You Dance: High-fidelity and Editable Dance Synthesis via Generative Masked Motion Prior
Shah, Foram N
Shah, Parshwa
Saleem, Muhammad Usama
Pinyoanuntapong, Ekkasit
Wang, Pu
Xue, Hongfei
Helmy, Ahmed
Graphics
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
Recent advances in dance generation have enabled the automatic synthesis of 3D dance motions. However, existing methods still face significant challenges in simultaneously achieving high realism, precise dance-music synchronization, diverse motion expression, and physical plausibility. To address these limitations, we propose a novel approach that leverages a generative masked text-to-motion model as a distribution prior to learn a probabilistic mapping from diverse guidance signals, including music, genre, and pose, into high-quality dance motion sequences. Our framework also supports semantic motion editing, such as motion inpainting and body part modification. Specifically, we introduce a multi-tower masked motion model that integrates a text-conditioned masked motion backbone with two parallel, modality-specific branches: a music-guidance tower and a pose-guidance tower. The model is trained using synchronized and progressive masked training, which allows effective infusion of the pretrained text-to-motion prior into the dance synthesis process while enabling each guidance branch to optimize independently through its own loss function, mitigating gradient interference. During inference, we introduce classifier-free logits guidance and pose-guided token optimization to strengthen the influence of music, genre, and pose signals. Extensive experiments demonstrate that our method sets a new state of the art in dance generation, significantly advancing both the quality and editability over existing approaches. Project Page available at https://foram-s1.github.io/DanceMosaic/
title Walk Before You Dance: High-fidelity and Editable Dance Synthesis via Generative Masked Motion Prior
topic Graphics
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2504.04634