Research map / Detection and segmentation / Open-set and few-shot detection
Neural-Logic Human-Object Interaction Detection
Liulei Li, Jianan Wei, Wenguan Wang, Yi Yang
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 triplet-reasoning attention to baseline (ablation) | baseline interaction decoder · HICO-DET Rare | mAP: +4.53 | Improved |
| add logic-guided reasoner learning to baseline with self-attention (ablation) | baseline · HICO-DET | mAP | Improved |
| combine TRA and logic-guided learning (full LogicHOI) (ablation) | LogicHOI · HICO-DET | mAP | Improved |
| add action-position loss L_v,p and object-position loss L_o,p (ablation) | LogicHOI · HICO-DET unseen verb/object zero-shot | mAP | Mixed |
| increase decoder layers from 3 to 6 (ablation) | LogicHOI · HICO-DET Full | mAP: 35.47 -> 35.61 | No clear effect |
| reduce decoder layers from 3 to 2 (ablation) | LogicHOI · HICO-DET Full | mAP: 35.47 -> 34.61 | Hurt |
| increase queries from 32 to 128 (ablation) | LogicHOI · HICO-DET Full | mAP: 35.47 -> 34.67 | Hurt |
| reduce queries from 32 to 16 (ablation) | LogicHOI · HICO-DET Full | mAP: 35.47 -> 35.06 | Hurt |
| LogicHOI vs GEN-VLKT | LogicHOI · HICO-DET unseen object | mAP: 10.51 -> 15.67 | Improved |
| LogicHOI vs GEN-VLKT | LogicHOI · HICO-DET RF-UC unseen | mAP: 21.36 -> 25.97 | Improved |
| LogicHOI vs STIP | LogicHOI · V-COCO scenario 1 | AP role: 65.1 -> 64.4 | Hurt |
| LogicHOI vs GEN-VLKT | LogicHOI · HICO-DET Default Full | mAP: 33.75 -> 35.47 | Improved |
Closest papers by meaning
- Detecting Any Human-Object Interaction Relationship: Universal HOI Detector with Spatial Prompt Learning on Foundation Models (NeurIPS 2023 · similarity 0.91)
- HOI Analysis: Integrating and Decomposing Human-Object Interaction (NeurIPS 2020 · similarity 0.90)
- HORP: Human-Object Relation Priors Guided HOI Detection (CVPR 2025 · similarity 0.89)
- Open-World Human-Object Interaction Detection via Multi-modal Prompts (CVPR 2024 · similarity 0.88)
- Exploring Pose-Aware Human-Object Interaction via Hybrid Learning (CVPR 2024 · similarity 0.88)
- Discovering Syntactic Interaction Clues for Human-Object Interaction Detection (CVPR 2024 · 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.
