Research map / 3D vision / 3D generation
MagicHOI: Leveraging 3D Priors for Accurate Hand-object Reconstruction from Short Monocular Video Clips
Shibo Wang, Haonan He, Maria Parelli, Christoph Gebhardt, Zicong Fan, Jie Song
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 NVS diffusion prior to RGB-only reconstruction (ablation) | MagicHOI · HO3D-v3 | Chamfer distance (cm): 1.37 -> 0.91 | Improved |
| add RGB observations to NVS-only reconstruction (ablation) | MagicHOI · HO3D-v3 | Chamfer distance (cm): 1.54 -> 0.91 | Improved |
| add visibility-aware weighting to RGB+NVS (ablation) | MagicHOI · HO3D-v3 | Chamfer distance (cm): 0.91 -> 0.89 | Improved |
| add visibility-aware weighting to RGB+NVS (ablation) | MagicHOI · HO3D-v3 | F10 (%): 90.81 -> 92.15 | Improved |
| add visibility-aware weighting to RGB+NVS (ablation) | MagicHOI · HO3D-v3 | hand-relative Chamfer distance (cm): 3.77 -> 2.39 | Improved |
| add NVS prior to RGB-only reconstruction (ablation) | MagicHOI · HO3D-v3 | hand-relative Chamfer distance (cm): 3.21 -> 3.77 | Hurt |
| add visibility-aware weighting to RGB+NVS (ablation) | MagicHOI · HO3D-v3 | Relative Scale: 0.17 -> 0.11 | Improved |
| optimize hand pose theta during alignment (ablation) | MagicHOI · HO3D-v3 | MPJPE (mm): 4.62 -> 7.83 | Hurt |
| use MagicHOI instead of HOLD | MagicHOI · HO3D-v3 | Chamfer distance (cm): 1.31 -> 0.87 | Improved |
| use MagicHOI instead of HOLD | MagicHOI · HO3D-v3 | MPJPE (mm): 30.79 -> 4.62 | Improved |
| use MagicHOI instead of EasyHOI | MagicHOI · HO3D-v3 | hand-relative Chamfer distance (cm): 19.55 -> 2.39 | Improved |
Later papers that compared with this method
1 later accepted paper reports MagicHOI as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- ForeHOI: Feed-forward 3D Object Reconstruction from Daily Hand-Object Interaction Videos (CVPR 2026 · better on 4 of 4 shared results)
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.
- HOLD — HOLD: Category-agnostic 3D Reconstruction of Interacting Hands and Objects from Video (CVPR 2024) · better on 6 of 6 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
- HOLD: Category-agnostic 3D Reconstruction of Interacting Hands and Objects from Video (CVPR 2024 · similarity 0.92)
- HandBooster: Boosting 3D Hand-Mesh Reconstruction by Conditional Synthesis and Sampling of Hand-Object Interactions (CVPR 2024 · similarity 0.91)
- EasyHOI: Unleashing the Power of Large Models for Reconstructing Hand-Object Interactions in the Wild (CVPR 2025 · similarity 0.91)
- BIGS: Bimanual Category-agnostic Interaction Reconstruction from Monocular Videos via 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
- Single-view Image to Novel-view Generation for Hand-Object Interactions (AAAI 2025 · similarity 0.90)
- The Surprising Effectiveness of Video Diffusion Models for Hand Motion Reconstruction (NeurIPS 2026 · similarity 0.90)
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.
