Research map / Image generation / Autoregressive and latent generation
TCDiff++: An End-to-end Trajectory-Controllable Diffusion Model for Harmonious Music-Driven Group Choreography
Yuqin Dai, Wanlu Zhu, Ronghui Li, Xiu Ying Li, Zhenyu Zhang, Jun Li, Jian Yang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 Fusion Projection (ablation) | TCDiff++ (cumulative) · AIOZ-GDance long-duration (cumulative ablation) | GMR: 67.24 -> 50.12 | Improved |
| add Dancer Positioning Embedding (ablation) | TCDiff++ (cumulative) · AIOZ-GDance long-duration (cumulative ablation) | GMR: 50.12 -> 44.38 | Improved |
| add Dancer Positioning Embedding (ablation) | TCDiff++ (cumulative) · AIOZ-GDance long-duration (cumulative ablation) | PFC: 3.01 -> 3.08 | Hurt |
| add distance-consistency loss (ablation) | TCDiff++ (cumulative) · AIOZ-GDance long-duration (cumulative ablation) | GMR: 44.38 -> 25.76 | Improved |
| add distance-consistency loss (ablation) | TCDiff++ (cumulative) · AIOZ-GDance long-duration (cumulative ablation) | TIF: 0.28 -> 0.19 | Improved |
| add Swap Mode embedding (ablation) | TCDiff++ (cumulative) · AIOZ-GDance long-duration (cumulative ablation) | TIF: 0.19 -> 0.17 | Improved |
| add Long Group Diffusion Sampling (ablation) | TCDiff++ (cumulative) · AIOZ-GDance long-duration (cumulative ablation) | GMR: 23.93 -> 23.52 | Improved |
| add Footwork Adaptor (ablation) | TCDiff++ (cumulative) · AIOZ-GDance long-duration (cumulative ablation) | PFC: 2.48 -> 1.49 | Improved |
| replace Transformer with State Space Model sequence decoder (ablation) | TCDiff++ (cumulative) · AIOZ-GDance long-duration (cumulative ablation) | GMR: 23.95 -> 14.67 | Improved |
| replace Transformer with State Space Model sequence decoder (ablation) | TCDiff++ (cumulative) · AIOZ-GDance long-duration (cumulative ablation) | FID: 24.75 -> 20.37 | Improved |
| replace Transformer with State Space Model sequence decoder (ablation) | TCDiff++ (cumulative) · AIOZ-GDance long-duration (cumulative ablation) | PFC: 1.49 -> 1.53 | Hurt |
| use end-to-end generation instead of two-stage trajectory-then-motion (TCDiff) | TCDiff++ · AIOZ-GDance (long-duration, 720 frames) | GMR: 23.87 -> 14.67 | Improved |
Closest papers by meaning
- Harmonious Music-driven Group Choreography with Trajectory-Controllable Diffusion (AAAI 2025 · similarity 0.95)
- Breaking the Group Size Barrier: Parameter-Efficient Group Dance Generation with Chain-of-Dancers (NeurIPS 2026 · similarity 0.91)
- Lodge: A Coarse to Fine Diffusion Network for Long Dance Generation Guided by the Characteristic Dance Primitives (CVPR 2024 · similarity 0.89)
- Bidirectional Autoregessive Diffusion Model for Dance Generation (CVPR 2024 · similarity 0.89)
- DanceAnyWay: Synthesizing Beat-Guided 3D Dances with Randomized Temporal Contrastive Learning (AAAI 2024 · similarity 0.88)
- MEGADance: Mixture-of-Experts Architecture for Genre-Aware 3D Dance Generation (NeurIPS 2025 · similarity 0.88)
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.
