Research map / 3D vision / 3D generation
ReMoGen: Real-time Human Interaction-to-Reaction Generation via Modular Learning from Diverse Data
Yaoqin Ye, Yiteng Xu, Qin Sun, Xinge Zhu, Yujing Sun, Yuexin Ma
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 |
|---|---|---|---|
| keep universal prior frozen and add Meta-Interaction modules instead of training from scratch (ablation) | ReMoGen · Inter-X (HHI) | FID: 0.270 -> 0.181 | Improved |
| keep prior frozen instead of joint finetuning the whole model (ablation) | ReMoGen · Inter-X (HHI) | FID: 0.298 -> 0.181 | Improved |
| add Meta-Interaction modules to the frozen prior (ablation) | ReMoGen · Inter-X (HHI) | FID: 3.735 -> 0.181 | Improved |
| add Frame-wise Segment Refinement at inference (ablation) | ReMoGen · Inter-X (HHI) | FID: 0.181 -> 0.166 | Improved |
| add FWSR at inference (ablation) | ReMoGen · Inter-X (HHI) | Latency (s/frame): 0.042 -> 0.047 | Hurt |
| use naive frame-wise sliding diffusion instead of segment rollout (ablation) | ReMoGen · Inter-X (HHI) | Latency (s/frame): 0.042 -> 0.305 | Hurt |
| use sliding inference with segment-trained model (ablation) | ReMoGen · Inter-X (HHI) | FID: 0.181 -> 1.785 | Hurt |
| compose HHI and HSI modules with equal weights (0.5 each) zero-shot (ablation) | ReMoGen · EgoBody (mixed, zero-shot) | FID: 15.087 -> 14.740 | Improved |
| initialize from universal prior before finetuning on EgoBody (ablation) | ReMoGen · EgoBody (finetune) | FID: 2.341 -> 0.418 | Improved |
| use HHI or HSI module alone zero-shot on mixed data | ReMoGen · EgoBody (mixed, zero-shot) | FID: 18.377 and 15.087 vs GT 0.000 | Hurt |
Closest papers by meaning
- ReGenNet: Towards Human Action-Reaction Synthesis (CVPR 2024 · similarity 0.90)
- InterDreamer: Zero-Shot Text to 3D Dynamic Human-Object Interaction (NeurIPS 2024 · similarity 0.89)
- TIMotion: Temporal and Interactive Framework for Efficient Human-Human Motion Generation (CVPR 2025 · similarity 0.89)
- Ready-to-React: Online Reaction Policy for Two-Character Interaction Generation (ICLR 2025 · similarity 0.88)
- InterAct: Advancing Large-Scale Versatile 3D Human-Object Interaction Generation (CVPR 2025 · similarity 0.88)
- Learning to Generate Human-Human-Object Interactions from Textual Descriptions (NeurIPS 2025 · similarity 0.87)
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.
