Salvato in:
| Autori principali: | , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2505.05589 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915836188426240 |
|---|---|
| author | Lin, Jingzhong Li, Xinru Qi, Yuanyuan Zhang, Bohao Liu, Wenxiang Tang, Kecheng Huang, Wenxuan Xu, Xiangfeng Li, Bangyan Wang, Changbo He, Gaoqi |
| author_facet | Lin, Jingzhong Li, Xinru Qi, Yuanyuan Zhang, Bohao Liu, Wenxiang Tang, Kecheng Huang, Wenxuan Xu, Xiangfeng Li, Bangyan Wang, Changbo He, Gaoqi |
| contents | Reactive dance generation (RDG), the task of generating a dance conditioned on a lead dancer's motion, holds significant promise for enhancing human-robot interaction and immersive digital entertainment. Despite progress in duet synchronization and motion-music alignment, two key challenges remain: generating fine-grained spatial interactions and ensuring long-term temporal coherence. In this work, we introduce \textbf{ReactDance}, a diffusion framework that operates on a novel hierarchical latent space to address these spatiotemporal challenges in RDG. First, for high-fidelity spatial expression and fine-grained control, we propose Hierarchical Finite Scalar Quantization (\textbf{HFSQ}). This multi-scale motion representation effectively disentangles coarse body posture from subtle limb dynamics, enabling independent and detailed control over both aspects through a layered guidance mechanism. Second, to efficiently generate long sequences with high temporal coherence, we propose Blockwise Local Context (\textbf{BLC}), a non-autoregressive sampling strategy. Departing from slow, frame-by-frame generation, BLC partitions the sequence into blocks and synthesizes them in parallel via periodic causal masking and positional encodings. Coherence across these blocks is ensured by a dense sliding-window training approach that enriches the representation with local temporal context. Extensive experiments show that ReactDance substantially outperforms state-of-the-art methods in motion quality, long-term coherence, and sampling efficiency. Project page: https://ripemangobox.github.io/ReactDance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_05589 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ReactDance: Hierarchical Representation for High-Fidelity and Coherent Long-Form Reactive Dance Generation Lin, Jingzhong Li, Xinru Qi, Yuanyuan Zhang, Bohao Liu, Wenxiang Tang, Kecheng Huang, Wenxuan Xu, Xiangfeng Li, Bangyan Wang, Changbo He, Gaoqi Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Reactive dance generation (RDG), the task of generating a dance conditioned on a lead dancer's motion, holds significant promise for enhancing human-robot interaction and immersive digital entertainment. Despite progress in duet synchronization and motion-music alignment, two key challenges remain: generating fine-grained spatial interactions and ensuring long-term temporal coherence. In this work, we introduce \textbf{ReactDance}, a diffusion framework that operates on a novel hierarchical latent space to address these spatiotemporal challenges in RDG. First, for high-fidelity spatial expression and fine-grained control, we propose Hierarchical Finite Scalar Quantization (\textbf{HFSQ}). This multi-scale motion representation effectively disentangles coarse body posture from subtle limb dynamics, enabling independent and detailed control over both aspects through a layered guidance mechanism. Second, to efficiently generate long sequences with high temporal coherence, we propose Blockwise Local Context (\textbf{BLC}), a non-autoregressive sampling strategy. Departing from slow, frame-by-frame generation, BLC partitions the sequence into blocks and synthesizes them in parallel via periodic causal masking and positional encodings. Coherence across these blocks is ensured by a dense sliding-window training approach that enriches the representation with local temporal context. Extensive experiments show that ReactDance substantially outperforms state-of-the-art methods in motion quality, long-term coherence, and sampling efficiency. Project page: https://ripemangobox.github.io/ReactDance. |
| title | ReactDance: Hierarchical Representation for High-Fidelity and Coherent Long-Form Reactive Dance Generation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2505.05589 |