Research map / Detection and segmentation / Open-set and few-shot detection
Detecting Human-Object Interactions via Functional Generalization
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 functional-generalization training augmentation (ablation) | HOI detector on HICO-Det · HICO-Det Full | mAP: 12.72 -> 14.35 | Helped |
| add functional-generalization training augmentation (ablation) | HOI detector on HICO-Det · HICO-Det Rare | mAP: 7.57 -> 9.84 | Helped |
| add 7 rather than 5 neighboring objects per training instance (ablation) | HOI detector on HICO-Det · HICO-Det Full | mAP: 14.35 -> 13.51 | Hurt |
| replace K-means with agglomerative clustering (ablation) | functional-generalization object clustering · HICO-Det Rare | mAP: 9.84 -> 7.59 | Hurt |
| replace K-means with Affinity Propagation (ablation) | functional-generalization object clustering · HICO-Det Full | mAP: 14.35 -> 13.49 | Hurt |
| add human visual feature f_h (ablation) | functional-generalization module · HICO-Det Full | mAP: 12.15 -> 14.35 | Helped |
| add geometric feature f_g (ablation) | functional-generalization module · HICO-Det Full | mAP: 12.43 -> 14.35 | Helped |
| add semantic word vectors w_h and w_o (ablation) | functional-generalization module · HICO-Det Full | mAP: 12.23 -> 14.35 | Helped |
| use functional generalization for seen-object zero-shot HOI detection | HOI detector · HICO-Det seen-object zero-shot, unseen 120 | mAP: 5.62 -> 11.31±1.03 | Helped |
| use Open Images generic object detector to support unseen objects | HOI detector · HICO-Det unseen-object zero-shot, unseen 100 | mAP | Helped |
Closest papers by meaning
- Learning Human-Object Interaction as Groups (NeurIPS 2025 · similarity 0.93)
- Exploring Pose-Aware Human-Object Interaction via Hybrid Learning (CVPR 2024 · similarity 0.90)
- HORP: Human-Object Relation Priors Guided HOI Detection (CVPR 2025 · similarity 0.90)
- HOI-M^3: Capture Multiple Humans and Objects Interaction within Contextual Environment (CVPR 2024 · similarity 0.90)
- Locality-Aware Zero-Shot Human-Object Interaction Detection (CVPR 2025 · similarity 0.90)
- An Image-like Diffusion Method for Human-Object Interaction Detection (CVPR 2025 · similarity 0.90)
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.