Research map / Detection and segmentation / Open-set and few-shot detection
Deep Structured Learning for Visual Relationship Detection
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 object features to predicate prediction (F -> Fo) (ablation) | feature-level model · VRD relationship detection | R@50: 12.96 -> 16.15 | Helped |
| add object features to predicate prediction (F -> Fo) (ablation) | feature-level model · VG relationship detection | R@50: 3.90 -> 4.77 | Helped |
| add label-level prediction (F -> F+L) (ablation) | F · VRD relationship detection | R@50: 12.96 -> 16.28 | Helped |
| add label-level prediction (F -> F+L) (ablation) | F · VG phrase prediction | precision: 16.58 -> 26.67 | Helped |
| add label-level prediction (Fo -> Fo+L) (ablation) | Fo · VRD relationship detection | R@50: 16.15 -> 17.27 | Helped |
| add label-level prediction (Fo -> Fo+L) (ablation) | Fo · VG relationship detection | R@50: 4.77 -> 6.37 | Helped |
| add label-level prediction (Fo -> Fo+L) (ablation) | Fo · VRD phrase prediction | precision: 35.94 -> 36.28 | No clear effect |
| add mask of training relationship types | Fo+L · VRD relationship detection | R@50: 17.27 -> 17.40 | Helped |
| add mask of training relationship types | Fo+L · VG relationship detection | R@50: 6.37 -> 6.82 | Helped |
| add mask of training relationship types | F · VRD relationship detection | R@50: 12.96 -> 14.85 | Helped |
| Fo+L versus RDN on relationship detection | Fo+L · VRD relationship detection | R@100: 20.88 -> 18.26 | Hurt |
Closest papers by meaning
- Visual Relationship Detection With Deep Structural Ranking (AAAI 2018 · similarity 0.92)
- Detecting Visual Relationships With Deep Relational Networks (CVPR 2017 · similarity 0.90)
- Tensorize, Factorize and Regularize: Robust Visual Relationship Learning (CVPR 2018 · similarity 0.89)
- Large-Scale Visual Relationship Understanding (AAAI 2019 · similarity 0.89)
- Groupwise Query Specialization and Quality-Aware Multi-Assignment for Transformer-based Visual Relationship Detection (CVPR 2024 · similarity 0.87)
- Multi-Modal Prompting for Open-Vocabulary Video Visual Relationship Detection (AAAI 2024 · similarity 0.85)
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.