Research map / Detection and segmentation / CNN features
RON: Reverse Connection With Objectness Prior Networks for Object Detection
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 |
|---|---|---|---|
| Use the full multi-scale testing, box voting, and flipping setup | RON384 on PASCAL VOC 2007 · PASCAL VOC 2007 test | mAP: 75.4 -> 77.6 | Helped |
| Use the full multi-scale testing, box voting, and flipping setup | RON384 on MS COCO · MS COCO test-dev2015 | AP 0.5:0.95: 25.4 -> 27.4 | Helped |
| Add detection from layers 5, 6, and 7 to layer 7 detection (ablation) | RON320 · PASCAL VOC 2007 test | mAP: 65.6 -> 72.5 | Helped |
| Add detection from layer 4 to detection from layers 5, 6, and 7 (ablation) | RON320 · PASCAL VOC 2007 test | mAP: 72.5 -> 74.2 | Helped |
| Add the objectness prior module (ablation) | RON320 · PASCAL VOC 2007 test | mAP: 69.6 -> 74.2 | Helped |
| Use MS COCO pretraining before fine-tuning on VOC | RON384++ · PASCAL VOC 2007 test | mAP: 77.6 -> 81.3 | Helped |
| Use MS COCO pretraining before fine-tuning on VOC | RON384++ · PASCAL VOC 2012 test | mAP: 75.4 -> 80.7 | Helped |
| Use RON rather than Faster R-CNN | Object detection on MS COCO · MS COCO test-dev2015 | AP 0.5:0.95: 21.9 -> 27.4 | Helped |
| Use RON384 rather than SSD500 | Object detection on PASCAL VOC 2007 · PASCAL VOC 2007 test | mAP: 75.1 -> 75.4 | Helped |
| Use top 10 RON region proposals rather than Faster R-CNN proposals | Region proposal generation · PASCAL VOC 2007 test | recall: 80.7%; 20 points higher | Helped |
Closest papers by meaning
- R-FCN: Object Detection via Region-based Fully Convolutional Networks (NeurIPS 2016 · similarity 0.90)
- Cascade R-CNN: Delving Into High Quality Object Detection (CVPR 2018 · similarity 0.89)
- Sparse R-CNN: End-to-End Object Detection With Learnable Proposals (CVPR 2021 · similarity 0.89)
- Spatially Invariant Unsupervised Object Detection with Convolutional Neural Networks (AAAI 2019 · similarity 0.89)
- ObjectNet: A large-scale bias-controlled dataset for pushing the limits of object recognition models (NeurIPS 2019 · similarity 0.89)
- HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection (CVPR 2016 · 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.