Research map / Detection and segmentation / CNN features
Glance and Gaze: Inferring Action-Aware Points for One-Stage Human-Object Interaction Detection
Xubin Zhong, Xian Qu, Changxing Ding, Dacheng Tao
Check 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 HNA loss (ablation) | GGNet · V-COCO | mAP: 51.06 -> 52.51 | Improved |
| Add HNA loss (ablation) | GGNet · HICO-DET DT | mAP: 20.16 -> 20.58 | Improved |
| Add Gaze Step 1 (ablation) | GGNet with HNA loss · V-COCO | mAP: 52.51 -> 54.03 | Improved |
| Add Gaze Step 1 (ablation) | GGNet with HNA loss · HICO-DET DT | mAP: 20.58 -> 21.05 | Improved |
| Add Gaze Step 2 (ablation) | GGNet with HNA loss and Gaze Step 1 · V-COCO | mAP: 54.03 -> 54.43 | Improved |
| Add Gaze Step 2 (ablation) | GGNet with HNA loss and Gaze Step 1 · HICO-DET DT | mAP: 21.05 -> 21.54 | Improved |
| Add APM (ablation) | GGNet with HNA loss and both gaze steps · V-COCO | mAP: 54.43 -> 54.72 | Improved |
| Add APM (ablation) | GGNet with HNA loss and both gaze steps · HICO-DET DT | mAP: 21.54 -> 22.03 | Improved |
| Increase ActPoints from 9 to 25 (ablation) | GGNet with 1 gaze step · V-COCO | AP_role: 53.54 -> 54.32 | Improved |
| Increase ActPoints from 25 to 49 (ablation) | GGNet with 1 gaze step · V-COCO | AP_role: 54.32 -> 53.81 | Hurt |
| Increase gaze steps from 1 to 2 (ablation) | GGNet with 25 ActPoints · V-COCO | AP_role: 54.21 -> 54.72 | Improved |
| Increase gaze steps from 2 to 3 (ablation) | GGNet with 25 ActPoints · V-COCO | AP_role: 54.72 -> 54.34 | Hurt |
Closest papers by meaning
- Exploring Pose-Aware Human-Object Interaction via Hybrid Learning (CVPR 2024 · similarity 0.88)
- Learning Human-Object Interaction as Groups (NeurIPS 2025 · similarity 0.88)
- ContextHOI: Spatial Context Learning for Human-Object Interaction Detection (AAAI 2025 · similarity 0.88)
- Exploring Self- and Cross-Triplet Correlations for Human-Object Interaction Detection (AAAI 2024 · similarity 0.87)
- HORP: Human-Object Relation Priors Guided HOI Detection (CVPR 2025 · similarity 0.87)
- Bilateral Adaptation for Human-Object Interaction Detection with Occlusion-Robustness (CVPR 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.
