Research map / Video / Video generation
You Never Stop Dancing: Non-freezing Dance Generation via Bank-constrained Manifold Projection
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add RefineBank (ablation) | FACT · AIST++ | FIDk: 35.35 -> 28.67 | Helped |
| add RefineBank (ablation) | FACT · AIST++ | Freezing rate: 39.0% -> 32.1% | Helped |
| add TransitBank (ablation) | FACT · AIST++ | FIDk: 35.35 -> 31.24 | Helped |
| add TransitBank (ablation) | FACT · AIST++ | Distk: 5.94 -> 7.42 | Helped |
| add both RefineBank and TransitBank (ablation) | FACT · AIST++ | FIDk: 35.35 -> 25.96 | Helped |
| replace vanilla AE with Bank-AE (ablation) | manifold projection · AIST++ | FIDk: 29.85 -> 25.96 | Helped |
| replace VAE with Bank-AE (ablation) | manifold projection · AIST++ | FIDk: 29.67 -> 25.96 | Helped |
| use convex combination of bank items instead of discrete nearest item (ablation) | Bank-AE · AIST++ | FIDk: 27.48 -> 25.96 | Helped |
| increase number of latent features n from 1 to 4 (ablation) | Bank-AE · AIST++ | FIDk: 27.94 -> 25.96 | Helped |
| increase number of latent features n from 4 to 5 (ablation) | Bank-AE · AIST++ | FIDk: 25.96 -> 26.67 | Hurt |
| increase bank items N from 32 to 256 (ablation) | Bank-AE · AIST++ | FIDk: 29.74 -> 25.96 | Helped |
| increase bank items N from 256 to 512 (ablation) | Bank-AE · AIST++ | FIDk: 25.96 -> 26.39 | Hurt |
Closest papers by meaning
- DanceAnyWay: Synthesizing Beat-Guided 3D Dances with Randomized Temporal Contrastive Learning (AAAI 2024 · similarity 0.88)
- Bidirectional Autoregessive Diffusion Model for Dance Generation (CVPR 2024 · similarity 0.87)
- MEGADance: Mixture-of-Experts Architecture for Genre-Aware 3D Dance Generation (NeurIPS 2025 · similarity 0.87)
- Scalable Motion Style Transfer with Constrained Diffusion Generation (AAAI 2024 · similarity 0.86)
- Breaking the Group Size Barrier: Parameter-Efficient Group Dance Generation with Chain-of-Dancers (NeurIPS 2026 · similarity 0.86)
- DanceCamera3D: 3D Camera Movement Synthesis with Music and Dance (CVPR 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.