Research map / Detection and segmentation / CNN features
Detecting Hands and Recognizing Physical Contact in the Wild
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 the Contact-Estimation module that reasons over surrounding objects | Mask R-CNN · 100DOH test (train: 100DOH) | mAP: 54.78 % -> 58.10 % | Helped |
| add the Contact-Estimation module that reasons over surrounding objects | Mask R-CNN · ContactHands test (train: ContactHands) | mAP: 53.31 % -> 57.41 % | Helped |
| train on ContactHands rather than 100DOH | Proposed model · ContactHands test | mAP: 33.37 % -> 57.41 % | Helped |
| train on ContactHands rather than 100DOH | Proposed model · 100DOH test | mAP: 33.37 % -> 41.76 % | Helped |
| combine 100DOH and ContactHands training data | Proposed model · 100DOH test | mAP: 58.10 % -> 59.54 % | Helped |
| combine 100DOH and ContactHands training data | Proposed model · ContactHands test | mAP: 57.41 % -> 58.93 % | Helped |
| add cross-feature affinity-based attention (ablation) | Contact-Estimation branch · ContactHands test | mAP: 56.08% -> 57.41% | Helped |
| add spatial attention (ablation) | Contact-Estimation branch · ContactHands test | mAP: 55.91% -> 57.41% | Helped |
| add both cross-feature affinity-based attention and spatial attention (ablation) | Contact-Estimation branch · ContactHands test | mAP: 55.12% -> 57.41% | Helped |
| use expanded axis-parallel hand crops instead of exact axis-parallel crops | ResNet-101 hand-crop classifier · ContactHands test | mAP: 26.10 % -> 38.56 % | Helped |
Closest papers by meaning
- Whose Hands Are These? Hand Detection and Hand-Body Association in the Wild (CVPR 2022 · similarity 0.90)
- Scale Invariant Fully Convolutional Network: Detecting Hands Efficiently (AAAI 2019 · similarity 0.89)
- Harmonious Feature Learning for Interactive Hand-Object Pose Estimation (CVPR 2023 · similarity 0.88)
- Learning Dense Hand Contact Estimation from Imbalanced Data (NeurIPS 2025 · similarity 0.88)
- Understanding Human Hands in Contact at Internet Scale (CVPR 2020 · similarity 0.88)
- Learning Explicit Contact for Implicit Reconstruction of Hand-Held Objects from Monocular Images (AAAI 2024 · similarity 0.87)
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.