Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly Supervised Region Proposal Network and Object Detection
Peng Tang, Xinggang Wang, Angtian Wang, Yongluan Yan, Wenyu Liu, Junzhou Huang, Alan Yuille
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 proposal refinement stage after coarse proposal generation (ablation) | weakly supervised region proposal network · PASCAL VOC 2007 | mAP: 37.5 -> 45.3 | Improved |
| add proposal refinement stage after coarse proposal generation (ablation) | weakly supervised region proposal network · PASCAL VOC 2007 trainval | CorLoc: 57.3 -> 63.8 | Improved |
| use our proposals instead of Edge Boxes (ablation) | OICR-based WSOD network · PASCAL VOC 2007 | mAP: 41.6 -> 45.3 | Improved |
| use our proposals instead of Edge Boxes (ablation) | OICR-based WSOD network · PASCAL VOC 2007 trainval | CorLoc: 60.7 -> 63.8 | Improved |
| use our proposals instead of Selective Search (ablation) | OICR-based WSOD network · PASCAL VOC 2007 | mAP: 42.2 -> 45.3 | Improved |
| use our proposals instead of Selective Search (ablation) | OICR-based WSOD network · PASCAL VOC 2007 trainval | CorLoc: 60.9 -> 63.8 | Improved |
| use box-supervised RPN proposals instead of our weakly supervised proposals (ablation) | OICR-based WSOD network · PASCAL VOC 2007 | mAP: 45.3 -> 46.2 | Improved |
| ensemble our proposals with Selective Search proposals | Ours-VGG16 · PASCAL VOC 2007 | mAP: 45.3 -> 47.9 | Improved |
| train Fast RCNN on top-scoring proposals as pseudo labels | Ours-Ens. · PASCAL VOC 2007 | mAP: 47.9 -> 50.4 | Improved |
| train Fast RCNN on top-scoring proposals as pseudo labels | Ours-VGG16-Ens. · PASCAL VOC 2012 | mAP: 43.4 -> 45.7 | Improved |
| ensemble our proposals with Selective Search proposals | Ours-VGG16 · PASCAL VOC 2012 | mAP: 40.8 -> 43.4 | Improved |
Closest papers by meaning
- You Reap What You Sow: Using Videos to Generate High Precision Object Proposals for Weakly-Supervised Object Detection (CVPR 2019 · similarity 0.91)
- WSOD2: Learning Bottom-Up and Top-Down Objectness Distillation for Weakly-Supervised Object Detection (ICCV 2019 · similarity 0.91)
- PCL: Proposal Cluster Learning for Weakly Supervised Object Detection (TPAMI 2018 · similarity 0.90)
- UWSOD: Toward Fully-Supervised-Level Capacity Weakly Supervised Object Detection (NeurIPS 2020 · similarity 0.90)
- W2N: Switching from Weak Supervision to Noisy Supervision for Object Detection (ECCV 2022 · similarity 0.89)
- Weakly Supervised Open-Vocabulary Object Detection (AAAI 2024 · 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.
