Research map / 3D vision / Human pose estimation
Learning a Contact Potential Field for Modeling the Hand-Object Interaction
Lixin Yang, Xinyu Zhan, Kailin Li, Wenqiang Xu, Junming Zhang, Jiefeng Li, Cewu Lu
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 |
|---|---|---|---|
| replace CPF elasticity with vanilla nearest-vertex contact attraction (ablation) | MIHO · FHB | Hand MPVPE (mm): 24.01 -> 19.54 | Improved |
| replace CPF with ObMan contact losses (ablation) | MIHO · FHB | Hand MPVPE (mm): 22.15 -> 19.54 | Improved |
| use CPF instead of ObMan contact losses (speed) (ablation) | MIHO · FHB | time per iteration (ms): 103.41 -> 55.77 | Improved |
| use CPF instead of vanilla contact on disjointedness (ablation) | MIHO · FHB | Disjointedness distance (mm): 16.32 -> 22.41 | Hurt |
| add repulsive springs (ablation) | MIHO · FHB | Solid intersection volume (cm3): 13.76 -> 11.76 | Improved |
| add repulsive springs (ablation) | MIHO · FHB | Penetration depth (mm): 17.79 -> 16.92 | Improved |
| add repulsive springs (ablation) | MIHO · FHB | Disjointedness distance (mm): 20.27 -> 22.41 | Hurt |
| use full A-MANO pose instead of PCA pose (ablation) | MIHO · FHB | Hand MPVPE (mm): 23.32 -> 19.54 | Improved |
| use full A-MANO pose instead of PCA pose (ablation) | MIHO · FHB | Penetration depth (mm): 22.47 -> 16.92 | Improved |
| increase repulsive spring stiffness k_rpl from 1.0e-3 to 8.0e-3 (ablation) | MIHO · FHB | Solid intersection volume (cm3): 11.76 -> 4.27 | Mixed |
| increase k_rpl from 1.0e-3 to 8.0e-3 (ablation) | MIHO · FHB | Disjointedness distance (mm): 22.41 -> 40.79 | Hurt |
| jointly optimize hand and object instead of hand only | MIHO · FHB | Hand MPVPE (mm): 21.16 -> 19.54 | Improved |
Closest papers by meaning
- CPF: Learning a Contact Potential Field To Model the Hand-Object Interaction (ICCV 2021 · similarity 0.98)
- Learning Explicit Contact for Implicit Reconstruction of Hand-Held Objects from Monocular Images (AAAI 2024 · similarity 0.88)
- EasyHOI: Unleashing the Power of Large Models for Reconstructing Hand-Object Interactions in the Wild (CVPR 2025 · similarity 0.88)
- VolCo: Volumetric Contact for High-Fidelity Human Grasp Generation (NeurIPS 2026 · similarity 0.88)
- Learning Dense Hand Contact Estimation from Imbalanced Data (NeurIPS 2025 · similarity 0.88)
- Single-view Image to Novel-view Generation for Hand-Object Interactions (AAAI 2025 · 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.
