Research map / Detection and segmentation / Open-set and few-shot detection
Universal-RCNN: Universal Object Detector via Transferable Graph R-CNN
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 |
|---|---|---|---|
| Use Universal-RCNN graph transfer and reasoning instead of multiple branches | FPN, ResNet-101 · MSCOCO minival | AP: 39.9 -> 44.7 | Helped |
| Use Universal-RCNN graph transfer and reasoning instead of multiple branches | FPN, ResNet-101 · VG | AP: 7.1 -> 8.8 | Helped |
| Use Universal-RCNN graph transfer and reasoning instead of multiple branches | FPN, ResNet-101 · ADE | AP: 12.2 -> 15.4 | Helped |
| Add intra-domain graph reasoning (ablation) | FPN on MSCOCO minival, trained with VG and COCO · MSCOCO minival | AP: 38.6 -> 41.4 | Helped |
| Add learned feature-based inter-domain transfer only (ablation) | FPN on MSCOCO minival, trained with VG and COCO · MSCOCO minival | AP: 38.6 -> 41.9 | Helped |
| Combine intra-domain reasoning with learned feature-based transfer (ablation) | FPN on MSCOCO minival, trained with VG and COCO · MSCOCO minival | AP: 38.6 -> 43.5 | Helped |
| Use learned feature-based transfer instead of attribute, relationship, or word-embedding transfer (ablation) | Universal-RCNN with intra-domain reasoning · MSCOCO minival | AP: 43.0; 43.2; 42.7 -> 43.5 | Helped |
| Transfer from VG using Universal-RCNN rather than multiple branches | FPN on MSCOCO minival · MSCOCO minival | mAP: 39.8 -> 43.5 | Helped |
| Add Universal-RCNN graph transfer and reasoning to FPN | FPN, ResNet-101 · MSCOCO test-val | AP: 38.9 -> 44.4 | Helped |
Closest papers by meaning
- Reasoning-RCNN: Unifying Adaptive Global Reasoning Into Large-Scale Object Detection (CVPR 2019 · similarity 0.90)
- Large Scale Semi-Supervised Object Detection Using Visual and Semantic Knowledge Transfer (CVPR 2016 · similarity 0.89)
- Towards Universal Object Detection by Domain Attention (CVPR 2019 · similarity 0.88)
- R-FCN: Object Detection via Region-based Fully Convolutional Networks (NeurIPS 2016 · similarity 0.88)
- SIGMA: Semantic-Complete Graph Matching for Domain Adaptive Object Detection (CVPR 2022 · similarity 0.88)
- Detecting Everything in the Open World: Towards Universal Object Detection (CVPR 2023 · similarity 0.87)
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.