Research map / Video / Video generation
Unified Multi-Modal Interactive and Reactive 3D Motion Generation via Rectified Flow
Prerit Gupta, Shourya Verma, Ananth Grama, Aniket Bera
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 RAG retrieval cross-attention (ablation) | DualFlow · MDD interactive (text+music) | FID: 0.622 -> 0.415 | Improved |
| add triplet contrastive loss (ablation) | DualFlow · MDD interactive (text+music) | FID: 0.783 -> 0.415 | Improved |
| add synchronization loss (ablation) | DualFlow · MDD interactive (text+music) | FID: 0.472 -> 0.415 | Improved |
| use Jukebox features instead of Mel-spectrogram (ablation) | DualFlow · MDD interactive (text+music) | FID: 0.647 -> 0.415 | Improved |
| add causal look-ahead attention (ablation) | DualFlow · MDD reactive (text+music) | FID: 0.849 -> 0.686 | Improved |
| add triplet contrastive loss (ablation) | DualFlow · MDD reactive (text+music) | FID: 0.885 -> 0.686 | Improved |
| add synchronization loss (ablation) | DualFlow · MDD reactive (text+music) | FID: 0.774 -> 0.686 | Improved |
| add RAG retrieval cross-attention (ablation) | DualFlow · MDD reactive (text+music) | FID: 0.714 -> 0.686 | Improved |
| add RAG retrieval cross-attention (ablation) | DualFlow · MDD reactive (text+music) | R-Precision Top-3 and MMDist: Top-3 0.479 -> 0.471; MMDist 0.933 -> 1.056 | Mixed |
| rectified flow with 20 steps vs InterGen 50 DDIM steps | DualFlow · MDD interactive (text+music) | inference time (s): 1.92 -> 1.24 | Improved |
| DualFlow vs InterGen on interactive FID | DualFlow · MDD interactive (text+music) | FID: 0.405 -> 0.415 | Hurt |
Closest papers by meaning
- Seamless Human Motion Composition with Blended Positional Encodings (CVPR 2024 · similarity 0.87)
- TIMotion: Temporal and Interactive Framework for Efficient Human-Human Motion Generation (CVPR 2025 · similarity 0.87)
- Everything2Motion: Synchronizing Diverse Inputs via a Unified Framework for Human Motion Synthesis (AAAI 2024 · similarity 0.87)
- AMD: Autoregressive Motion Diffusion (AAAI 2024 · similarity 0.87)
- InterDreamer: Zero-Shot Text to 3D Dynamic Human-Object Interaction (NeurIPS 2024 · similarity 0.86)
- BOTH2Hands: Inferring 3D Hands from Both Text Prompts and Body Dynamics (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.
