Research map / Detection and segmentation / Open-set and few-shot detection
Mining Cross-Person Cues for Body-Part Interactiveness Learning in HOI Detection
Xiaoqian Wu, Yong-Lu Li, Xinpeng Liu, Junyi Zhang, Yuzhe Wu, Cewu Lu
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 body-part level features (ablation) | Ours · HICO-DET | interactiveness AP: 36.46 -> 38.74 | Improved |
| add body-part saliency map (ablation) | Ours · HICO-DET | interactiveness AP: 37.43 -> 38.74 | Improved |
| add progressive masking (ablation) | Ours · HICO-DET | interactiveness AP: 38.06 -> 38.74 | Improved |
| add progressive masking (ablation) | Ours · HICO-DET | HOI mAP: 34.05 -> 35.15 | Improved |
| add sparsity adaptive sampling (ablation) | Ours · HICO-DET | interactiveness AP: 38.29 -> 38.74 | Improved |
| add one-time passing (filter and merge body-parts) (ablation) | Ours · HICO-DET | interactiveness AP: 38.51 -> 38.74 | Improved |
| add one-time passing (ablation) | Ours · HICO-DET | training time per epoch: 33 min/epoch -> 28 min/epoch | Improved |
| replace per-body-part passes with merged design (intuitive scheme vs full) (ablation) | Ours · HICO-DET | HOI mAP: 33.12 -> 35.15 | Improved |
| add body-part supervision from HAKE labels (ablation) | Ours · HICO-DET | body-part interactiveness AP: 1.38 -> 31.32 (arms) | Improved |
| randomly drop border tokens of body-part masks (ablation) | Ours · HICO-DET | interactiveness AP: 38.72 -> 38.74 | No clear effect |
| progressive masks versus one fixed mask on all layers (ablation) | Ours · HICO-DET | interactiveness AP: 37.91/38.06/37.95 -> 38.74 | Improved |
| use our detected pairs to feed iCAN instead of CDN pairs | iCAN · HICO-DET | mAP (Full): 21.09 -> 24.38 | Improved |
Closest papers by meaning
- Interactiveness Field in Human-Object Interactions (CVPR 2022 · similarity 0.90)
- Learning Human-Object Interaction as Groups (NeurIPS 2025 · similarity 0.89)
- Exploring Pose-Aware Human-Object Interaction via Hybrid Learning (CVPR 2024 · similarity 0.88)
- Exploring the Potential of Large Foundation Models for Open-Vocabulary HOI Detection (CVPR 2024 · similarity 0.88)
- HORP: Human-Object Relation Priors Guided HOI Detection (CVPR 2025 · similarity 0.88)
- Exploring Self- and Cross-Triplet Correlations for Human-Object Interaction Detection (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.
