Research map / 3D vision / 3D generation
ArtHOI: Taming Foundation Models for Monocular 4D Reconstruction of Hand-Articulated-Object Interactions
Zikai Wang, Zhilu Zhang, Yiqing Wang, Hui Li, Wangmeng Zuo
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 |
|---|---|---|---|
| use ArtHOI instead of RSRD (no pre-scan needed) | monocular articulated HOI reconstruction · ArtHOI-RGBD Scissor | Chamfer distance (mm): 13.841 -> 4.256 | Improved |
| use ArtHOI instead of RSRD | monocular articulated HOI reconstruction · RSRD dataset | Chamfer distance (mm): LED Light 10.144 -> 10.836; Sunglasses 31.985 -> 9.956 | Mixed |
| add Adaptive Sampling Refinement instead of FoundationPose (ablation) | metric scale and pose estimation · ArtHOI-RGBD | pose IoU: 0.820 -> 0.905 | Improved |
| add ASR instead of FoundationPose (ablation) | metric scale and pose estimation · ArtHOI-RGBD | success rate: 60% -> 100% | Improved |
| add ASR instead of Any6D (ablation) | metric scale and pose estimation · ArtHOI-Wild | pose IoU: 0.683 -> 0.882 | Improved |
| add MLLM-guided contact alignment (ablation) | ArtHOI · ArtHOI-RGBD | Co2 score: 0.972 -> 0.029 | Improved |
| use MLLM contact reasoning instead of mask-intersection heuristic (ablation) | ArtHOI alignment · ArtHOI-Wild | Co2 score: 0.059 -> 0.039 | Improved |
| add temporal, perspective, MinFP and depth prompting (ablation) | MLLM contact reasoning · ArtHOI-Wild | false positive rate (%): 16.59 -> 9.81 | Improved |
| add all prompting components (ablation) | MLLM contact reasoning · ArtHOI-Wild | accuracy (%): 83.50 -> 86.56 | Improved |
| use ArtHOI instead of RSRD | articulated reconstruction · ARCTIC subset | reconstruction | Improved |
Closest papers by meaning
- EasyHOI: Unleashing the Power of Large Models for Reconstructing Hand-Object Interactions in the Wild (CVPR 2025 · similarity 0.91)
- HOLD: Category-agnostic 3D Reconstruction of Interacting Hands and Objects from Video (CVPR 2024 · similarity 0.91)
- HandBooster: Boosting 3D Hand-Mesh Reconstruction by Conditional Synthesis and Sampling of Hand-Object Interactions (CVPR 2024 · similarity 0.89)
- Hand-held Object Reconstruction from RGB Video with Dynamic Interaction (CVPR 2025 · similarity 0.89)
- BIGS: Bimanual Category-agnostic Interaction Reconstruction from Monocular Videos via 3D Gaussian Splatting (CVPR 2025 · similarity 0.88)
- UniCon3R: Unified Contact-aware 4D Human-Scene Reconstruction from Monocular Video (NeurIPS 2026 · 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.
