Research map / 3D vision / Human pose estimation
Detecting Human-Object Contact in Images
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 body-part attention branch (ablation) | HOT contact detector · HOT-Annotated | SC-Acc.: 24.1 -> 40.7 | Helped |
| Add the body-part attention branch (ablation) | HOT contact detector · HOT-Generated | SC-Acc.: 12.0 -> 30.4 | Helped |
| Add the body-part attention branch (ablation) | HOT contact detector · HOT Full Set | SC-Acc.: 19.4 -> 36.4 | Helped |
| Add human-part supervision to the attention branch during early training (ablation) | HOT contact detector · HOT-Annotated | SC-Acc.: 33.8 -> 40.7 | Helped |
| Add human-part supervision to the attention branch during early training (ablation) | HOT contact detector · HOT-Generated | SC-Acc.: 20.3 -> 30.4 | Helped |
| Add human-part supervision to the attention branch during early training (ablation) | HOT contact detector · HOT Full Set | SC-Acc.: 30.4 -> 36.4 | Helped |
| Train using HOT-Annotated instead of HOT-Generated | HOT contact detector · HOT-Generated test | SC-Acc.: 21.1 -> 28.4 | Helped |
| Train using HOT-Generated instead of HOT-Annotated | HOT contact detector · HOT-Annotated test | SC-Acc.: 40.7 -> 2.46 | Hurt |
| Combine HOT-Generated with HOT-Annotated training data | HOT contact detector · HOT-Generated test | SC-Acc.: 30.4 -> 34.3 | Helped |
| Combine HOT-Generated with HOT-Annotated training data | HOT contact detector · HOT-Annotated test | SC-Acc.: 40.7 -> 47.4 | Helped |
| Use HOT-pGT in addition to RICH training data | BSTRO 3D contact estimator · HOT | f1: 0.231 -> 0.636 | Helped |
| Use HOT-pGT in addition to RICH training data | BSTRO 3D contact estimator · RICH | f1: 0.708 -> 0.684 | Hurt |
Closest papers by meaning
- PICO: Reconstructing 3D People In Contact with Objects (CVPR 2025 · similarity 0.90)
- Precision-Enhanced Human-Object Contact Detection via Depth-Aware Perspective Interaction and Object Texture Restoration (AAAI 2025 · similarity 0.89)
- Learning Explicit Contact for Implicit Reconstruction of Hand-Held Objects from Monocular Images (AAAI 2024 · similarity 0.88)
- ContactField: Implicit Field Representation for Multi-Person Interaction Geometry (NeurIPS 2024 · similarity 0.87)
- HOT3D: Hand and Object Tracking in 3D from Egocentric Multi-View Videos (CVPR 2025 · similarity 0.87)
- InteractVLM: 3D Interaction Reasoning from 2D Foundational Models (CVPR 2025 · 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.