EUVIMEDEUVIMED entdecken
Crossref

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

Foram N Shah, Parshwa N Shah, Muhammad Usama Saleem, Ekkasit Pinyoanuntapong, Pu Wang, Hongfei Xue, Ahmed Helmy

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (11) · 8796-8804

DOI: 10.1609/aaai.v40i11.37833 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Foram N Shah und 6 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

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 the quality and editability over existing approaches.

Zitieren

APA

Shah, F. N., Shah, P. N., Saleem, M. U., Pinyoanuntapong, E., Wang, P., Xue, H., & Helmy, A. (2026). Walk Before You Dance: High-fidelity and Editable Dance Synthesis via Generative Masked Motion Prior. Proceedings of the AAAI Conference on Artificial Intelligence, 40(11), 8796-8804. https://doi.org/10.1609/aaai.v40i11.37833

Vancouver

Shah FN, Shah PN, Saleem MU, Pinyoanuntapong E, Wang P, Xue H, et al. Walk Before You Dance: High-fidelity and Editable Dance Synthesis via Generative Masked Motion Prior. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(11):8796-8804. doi:10.1609/aaai.v40i11.37833

BibTeX
@article{Shah2026,
  author = {Shah, Foram N and Shah, Parshwa N and Saleem, Muhammad Usama and Pinyoanuntapong, Ekkasit and Wang, Pu and Xue, Hongfei and Helmy, Ahmed},
  title = {Walk Before You Dance: High-fidelity and Editable Dance Synthesis via Generative Masked Motion Prior},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {11},
  pages = {8796--8804},
  doi = {10.1609/aaai.v40i11.37833}
}
Kostenlos starten →