Research map / Detection and segmentation / Open-set and few-shot detection
SGAP-Net: Semantic-Guided Attentive Prototypes Network for Few-Shot Human-Object Interaction Recognition
Zhong Ji, Xiyao Liu, Yanwei Pang, Xuelong Li
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 verb semantic guidance only (ablation) | Prototypical Networks · HICO-FS test | 5-way 1-shot accuracy: 32.56 -> 35.86 | Improved |
| add noun semantic guidance only (ablation) | Prototypical Networks · HICO-FS test | 5-way 5-shot accuracy: 42.49 -> 54.68 | Improved |
| combine verb and noun linearly without attention-erasing (ablation) | SGAP-Net · HICO-FS test | 5-way 1-shot accuracy: 32.56 -> 32.03 | Hurt |
| add attention-erasing for verb and noun guidance (ablation) | SGAP-Net · HICO-FS test | 5-way 5-shot accuracy: 49.06 -> 58.39 | Improved |
| add PS-Module (prototype shift decision) (ablation) | SGAP-Net · HICO-FS test | 5-way 1-shot accuracy: 37.48 -> 38.16 | Improved |
| add PS-Module (prototype shift decision) (ablation) | SGAP-Net · HICO-FS test | 5-way 5-shot accuracy: 56.79 -> 58.39 | Improved |
| use PS-Module instead of Prototypical Net decision | 4-layer CNN few-shot · miniImageNet | 5-way accuracy: 44.32 -> 45.14 (1-shot); 63.73 -> 64.85 (5-shot) | Improved |
| use attention-erasing threshold tau 1.5 instead of 0 (ablation) | SGAP-Net · HICO-FS test | 5-way accuracy | Improved |
| use query-shift scaling alpha 0.5 (ablation) | SGAP-Net PS-Module · HICO-FS test | 5-way accuracy | Improved |
Closest papers by meaning
- Few-Shot Learning from Augmented Label-Uncertain Queries in Bongard-HOI (AAAI 2024 · similarity 0.90)
- Learning Transferable Human-Object Interaction Detector With Natural Language Supervision (CVPR 2022 · similarity 0.88)
- Toward Open-Set Human Object Interaction Detection (AAAI 2024 · similarity 0.87)
- Exploring the Potential of Large Foundation Models for Open-Vocabulary HOI Detection (CVPR 2024 · similarity 0.87)
- Visual Diversity and Region-aware Prompt Learning for Zero-shot HOI Detection (NeurIPS 2025 · similarity 0.87)
- Exploring Pose-Aware Human-Object Interaction via Hybrid Learning (CVPR 2024 · similarity 0.86)
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.
