Research map / Detection and segmentation / Open-set and few-shot detection
Prompt Guidance and Human Proximal Perception for HOT Prediction with Regional Joint Loss
Yuxiao Wang, Yu Lei, Zhenao Wei, Weiying Xue, Xinyu Jiang, Nan Zhuang, Qi Liu
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 Fine feature refinement in decoder (ablation) | baseline · HOT-Annotated | SC-Acc: 37.2 -> 38.9 | Improved |
| add text encoder (TE) to BF (ablation) | BF · HOT-Annotated | AD-Acc: 34.1 -> 37.1 | Improved |
| add depth model (DM) to BF (ablation) | BF · HOT-Annotated | AD-Acc: 34.1 -> 36.5 | Improved |
| add SAM to BF (ablation) | BF · HOT-Annotated | AD-Acc: 34.1 -> 35.3 | Improved |
| add DM and SAM (HPP) on top of BF+TE (ablation) | BF+TE · HOT-Annotated | SC-Acc: 40.3 -> 43.2 | Improved |
| add binary cross-entropy loss to CE (ablation) | P3HOT · HOT-Annotated | SC-Acc: 43.2 -> 44.5 | Improved |
| add RJLoss (ablation) | P3HOT · HOT-Annotated | SC-Acc: 44.5 -> 46.0 | Improved |
| normalize depth map D to [0,1] (ablation) | P3HOT · HOT-Annotated | SC-Acc: 44.1 -> 46.0 | Improved |
| lower loss weights alpha, beta from 1.0 to 0.3, 0.1 (ablation) | P3HOT · HOT-Annotated | SC-Acc: 42.3 -> 46.0 | Improved |
| lower gamma from 1.0 to 0.5 (ablation) | P3HOT · HOT-Annotated | SC-Acc: 46.0 -> 45.5 | Hurt |
| P3HOT versus PIHOT on C-Acc. | P3HOT · HOT-Annotated | C-Acc: 80.7 -> 74.9 | Hurt |
Closest papers by meaning
- Precision-Enhanced Human-Object Contact Detection via Depth-Aware Perspective Interaction and Object Texture Restoration (AAAI 2025 · similarity 0.88)
- Visual Diversity and Region-aware Prompt Learning for Zero-shot HOI Detection (NeurIPS 2025 · similarity 0.84)
- The 3D-PC: a benchmark for visual perspective taking in humans and machines (ICLR 2025 · similarity 0.84)
- PICO: Reconstructing 3D People In Contact with Objects (CVPR 2025 · similarity 0.84)
- InteractAnything: Zero-shot Human Object Interaction Synthesis via LLM Feedback and Object Affordance Parsing (CVPR 2025 · similarity 0.84)
- CG-HOI: Contact-Guided 3D Human-Object Interaction Generation (CVPR 2024 · similarity 0.83)
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.
