Research map / 3D vision / 3D generation
HAT-4D: Lifting Monocular Video for 4D Multi-Object Interactions via Human-Agent Collaboration
Jiaxin Li, Yuxiang Wu, Zhenkai Zhang, Xinrui Shi, wang haoyuan, Yichen Zhao, Su Linxiang, Chenyang chenyang, Mingyu Zhang, Yifan Ding, Boran Wen, li zhang and 2 more
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 Interaction Knowledge Graph (ablation) | HAT-4D · MVOIK-4D | Deform / Relation / Long: Deform 3.2835 -> 3.4487; Long 0.0708 -> 0.0901 | Improved |
| add Interaction Knowledge Graph (ablation) | HAT-4D · MVOIK-4D | CLIP / FVD: CLIP 0.8370 -> 0.8275; FVD 822.5620 -> 830.4060 | Hurt |
| add memory bank (ablation) | HAT-4D · MVOIK-4D | Long-term memory: 0.0621 -> 0.0901 | Improved |
| use HAT-4D instead of L4GM | HAT-4D · MVOIK-4D | FVD: 962.8593 -> 830.4060 | Improved |
| allow 3 human feedbacks instead of 0 (ablation) | HAT-4D · MVOIK-4D (39 sequences) | Deform: 3.9038 -> 5.4744 | Improved |
| allow 7 instead of 5 human feedbacks (ablation) | HAT-4D · MVOIK-4D (39 sequences) | Deform: 5.9295 -> 5.8974 | No clear effect |
| add object-level human refinement (ablation) | HAT-4D · MVOIK-4D (10 cases) | FVD: 860.5518 -> 685.7023 | Improved |
| add region-level refinement to object-level (ablation) | HAT-4D · MVOIK-4D (10 cases) | Deform: 6.1750 -> 5.8000 | Hurt |
| finetune L4GM with jittered views only (ablation) | L4GM · MVOIK-4D | PSNR | No clear effect |
| finetune L4GM with randomly sampled views (ablation) | L4GM · MVOIK-4D | PSNR | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- L4GM — L4GM: Large 4D Gaussian Reconstruction Model (NeurIPS 2024) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Real2Sim in HOI: Toward Physically Plausible HOI Reconstruction from Monocular Videos (NeurIPS 2026 · similarity 0.89)
- CORE4D: A 4D Human-Object-Human Interaction Dataset for Collaborative Object REarrangement (CVPR 2025 · similarity 0.87)
- UniCon3R: Unified Contact-aware 4D Human-Scene Reconstruction from Monocular Video (NeurIPS 2026 · similarity 0.87)
- Harmony4D: A Video Dataset for In-The-Wild Close Human Interactions (NeurIPS 2024 · similarity 0.86)
- CAT4D: Create Anything in 4D with Multi-View Video Diffusion Models (CVPR 2025 · similarity 0.86)
- Feature4X: Bridging Any Monocular Video to 4D Agentic AI with Versatile Gaussian Feature Fields (CVPR 2025 · 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.
