Research map / Detection and segmentation / Open-set and few-shot detection
Detecting Human-Object Interaction via Fabricated Compositional Learning
Check 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 fabricated object compositional learning | HICO-DET HOI detection · HICO-DET Default Rare | mAP: 17.08 -> 20.03 | Helped |
| Add fabricated object compositional learning | HICO-DET zero-shot HOI detection · Unseen composition, rare-first | mAP: 8.94 -> 13.16 | Helped |
| Add fabricated object compositional learning | HICO-DET zero-shot HOI detection · Unseen composition, non-rare-first | mAP: 13.47 -> 18.66 | Helped |
| Add fabricated object compositional learning | HICO-DET zero-shot HOI detection · Unseen object | mAP: 12.86 -> 15.54 | Helped |
| Add fabricated object compositional learning | PMFNet on V-COCO · V-COCO | AP_role: 51.85 -> 52.35 | Helped |
| Add noise to the object fabricator (ablation) | FCL · HICO-DET zero-shot ablation, Rare | mAP: 17.69 -> 18.69 | Helped |
| Provide verb representation to the object fabricator (ablation) | FCL · HICO-DET zero-shot ablation, Unseen | mAP: 14.71 -> 15.86 | Helped |
| Fabricate object features rather than verb features (ablation) | FCL · HICO-DET zero-shot ablation, Rare | mAP: 16.93 -> 18.69 | Helped |
| Add auxiliary verb regularization (ablation) | FCL · HICO-DET zero-shot ablation, Rare | mAP: 15.99 -> 18.69 | Helped |
| Replace step-wise training with one-step optimization (ablation) | FCL · HICO-DET zero-shot, Unseen | mAP: 15.86 -> 17.64 | Helped |
| Replace step-wise training with one-step optimization (ablation) | FCL · HICO-DET zero-shot, Rare | mAP: 18.22 -> 18.69 | Hurt |
| Use a fine-tuned HICO-DET detector instead of the COCO detector (ablation) | FCL · HICO-DET long-tailed, Rare | mAP: 18.73 -> 20.03 | Helped |
Closest papers by meaning
- Weakly-supervised HOI Detection via Prior-guided Bi-level Representation Learning (ICLR 2023 · similarity 0.91)
- Exploring Pose-Aware Human-Object Interaction via Hybrid Learning (CVPR 2024 · similarity 0.89)
- Human-Object Interaction Detection Collaborated with Large Relation-driven Diffusion Models (NeurIPS 2024 · similarity 0.89)
- Open-World Human-Object Interaction Detection via Multi-modal Prompts (CVPR 2024 · similarity 0.89)
- Toward Open-Set Human Object Interaction Detection (AAAI 2024 · similarity 0.89)
- Exploring the Potential of Large Foundation Models for Open-Vocabulary HOI Detection (CVPR 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.