Research map / Detection and segmentation / Open-set and few-shot detection
Visual Relationship Detection With Deep Structural Ranking
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 |
|---|---|---|---|
| Deep Structural Ranking vs DR-Net | relationship detection · VRD predicate det | R@100: 81.90 -> 93.18 | Helped |
| Deep Structural Ranking vs DR-Net | relationship detection · VRD relationship det | R@100: 20.88 -> 23.29 | Helped |
| add separate subject/object visual features to union box (ablation) | DSR · VRD predicate det | R@50: 66.14 -> 70.58 | Helped |
| add spatial mask cue (ablation) | DSR (V2) · VRD predicate det | R@50: 70.58 -> 78.14 | Helped |
| add word2vec semantic embedding cue (ablation) | DSR (V2) · VRD predicate det | R@50: 70.58 -> 84.38 | Helped |
| add spatial mask cue on top of visual+semantic (ablation) | DSR (V2E) · VRD predicate det | R@50: 84.38 -> 86.01 | Helped |
| spatial masks instead of coordinate features (ablation) | DSR (V2S1E) · VRD predicate det | R@50: 86.32 -> 86.01 | No clear effect |
| structural ranking loss instead of cross-entropy (ablation) | DSR (V2S2E) · VRD predicate det | R@50: 79.22 -> 86.01 | Helped |
| structural ranking loss instead of softmax (ablation) | DSR (V2S2E) · VRD relationship det | R@100: 22.09 -> 23.29 | Helped |
| increase alpha weight on training-set prior (ablation) | DSR inference · VRD zero-shot relationship det | recall | Hurt |
Closest papers by meaning
- Deep Structured Learning for Visual Relationship Detection (AAAI 2018 · similarity 0.92)
- Detecting Visual Relationships With Deep Relational Networks (CVPR 2017 · similarity 0.91)
- On Exploring Undetermined Relationships for Visual Relationship Detection (CVPR 2019 · similarity 0.91)
- Multi-Modal Prompting for Open-Vocabulary Video Visual Relationship Detection (AAAI 2024 · similarity 0.87)
- Groupwise Query Specialization and Quality-Aware Multi-Assignment for Transformer-based Visual Relationship Detection (CVPR 2024 · similarity 0.87)
- Towards Flexible Visual Relationship Segmentation (NeurIPS 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.