Research map / Detection and segmentation / Open-set and few-shot detection
Pose-Aware Multi-Level Feature Network for Human Object Interaction Detection
Bo Wan, Desen Zhou, Yongfei Liu, Rongjie Li, Xuming He
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 spatial configuration map (SCM) (ablation) | baseline · V-COCO val | AP_role: 49.2 -> 49.9 | Improved |
| add part-crop (PC) zoom-in part features (ablation) | baseline+SCM · V-COCO val | AP_role: 49.9 -> 51.0 | Improved |
| add spatial align (SpAlign) (ablation) | baseline+SCM+PC · V-COCO val | AP_role: 51.0 -> 52.4 | Improved |
| add semantic attention (SeAtten) (ablation) | baseline+SCM+PC+SpAlign · V-COCO val | AP_role: 52.4 -> 52.7 | Improved |
| add interaction affinity (IA) (ablation) | PMFNet without IA · V-COCO val | AP_role: 52.7 -> 53.0 | Improved |
| add SCM to full model (drop-one-out) (ablation) | PMFNet · V-COCO val | AP_role: 52.0 -> 53.0 | Improved |
| add PC, SpAlign, SeAtten together (drop-one-out) (ablation) | PMFNet · V-COCO val | AP_role: 50.3 -> 53.0 | Improved |
| add SpAlign to full model (drop-one-out) (ablation) | PMFNet · V-COCO val | AP_role: 51.1 -> 53.0 | Improved |
| add SeAtten to full model (drop-one-out) (ablation) | PMFNet · V-COCO val | AP_role: 52.6 -> 53.0 | Improved |
| add all pose-aware modules to baseline | Our baseline · V-COCO test | AP_role: 48.6 -> 52.0 | Improved |
| add all pose-aware modules to baseline | Our baseline · HICO-DET Default Full | mAP: 14.92 -> 17.46 | Improved |
| add all pose-aware modules to baseline | Our baseline · HICO-DET Know Object Full | mAP: 18.83 -> 20.34 | Improved |
Closest papers by meaning
- Learning Human-Object Interaction Detection Using Interaction Points (CVPR 2020 · similarity 0.92)
- Exploring Pose-Aware Human-Object Interaction via Hybrid Learning (CVPR 2024 · similarity 0.91)
- LINK: Learning Instance-level Knowledge from Vision-Language Models for Human-Object Interaction Detection (ICLR 2026 · similarity 0.91)
- InteractAnything: Zero-shot Human Object Interaction Synthesis via LLM Feedback and Object Affordance Parsing (CVPR 2025 · similarity 0.88)
- HORP: Human-Object Relation Priors Guided HOI Detection (CVPR 2025 · similarity 0.88)
- Human-Object Interaction Detection Collaborated with Large Relation-driven Diffusion Models (NeurIPS 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.
