Research map / 3D vision / Human pose estimation
What's in Your Hands? 3D Reconstruction of Generic Objects in Hands
Yufei Ye, Abhinav Gupta, Shubham Tulsiani
Check 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 |
|---|---|---|---|
| Condition object reconstruction on predicted hand articulation (ablation) | Ours · ObMan | F-5: 0.37 -> 0.42 | Improved |
| Condition object reconstruction on predicted hand articulation (ablation) | Ours · HO3D | F-5: 0.27 -> 0.33 | Improved |
| Condition object reconstruction on predicted hand articulation (ablation) | Ours · MOW | F-5: 0.10 -> 0.13 | Improved |
| Condition object reconstruction on predicted hand articulation (ablation) | Ours · ObMan | intersection volume: 3.93 -> 1.74 | Hurt |
| Condition object reconstruction on predicted hand articulation (ablation) | Ours · HO3D | intersection volume: 6.30 -> 4.77 | Improved |
| Condition object reconstruction on predicted hand articulation (ablation) | Ours · MOW | intersection volume: 17.3 -> 19.4 | Hurt |
| Use articulation-aware positional encoding instead of pose parameters (ablation) | Ours · ObMan | F5: 0.23 -> 0.42 | Improved |
| Use articulation-aware positional encoding instead of pose parameters with ground-truth hand pose (ablation) | Ours · ObMan | F5: 0.46 -> 0.49 | Improved |
| Use predicted hand articulation rather than ground-truth hand articulation (ablation) | Ours with articulation-aware encoding · ObMan | F5: 0.49 -> 0.42 | Hurt |
| Use predicted hand articulation rather than ground-truth hand articulation (ablation) | Ours with pose-parameter encoding · ObMan | F5: 0.46 -> 0.23 | Hurt |
| Apply test-time refinement of hand pose using object interaction constraints (ablation) | Ours · ObMan | CD: 1.02 -> 1.00 | Improved |
| Apply test-time refinement of hand pose using object interaction constraints (ablation) | Ours · ObMan | intersection volume: 1.74 -> 1.28 | Improved |
Closest papers by meaning
- HOLD: Category-agnostic 3D Reconstruction of Interacting Hands and Objects from Video (CVPR 2024 · similarity 0.93)
- Learning Explicit Contact for Implicit Reconstruction of Hand-Held Objects from Monocular Images (AAAI 2024 · similarity 0.92)
- Hand-held Object Reconstruction from RGB Video with Dynamic Interaction (CVPR 2025 · similarity 0.91)
- EasyHOI: Unleashing the Power of Large Models for Reconstructing Hand-Object Interactions in the Wild (CVPR 2025 · similarity 0.90)
- ComPose: When to Trust Hands for Object Pose Tracking (NeurIPS 2026 · similarity 0.89)
- SRHand: Super-Resolving Hand Images and 3D Shapes via View/Pose-aware Neural Image Representations and Explicit Meshes (NeurIPS 2025 · similarity 0.89)
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.
