Research map / Detection and segmentation / Weakly supervised detection and segmentation
You Reap What You Sow: Using Videos to Generate High Precision Object Proposals for Weakly-Supervised 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 video-trained W-RPN proposals to rank high-recall edge box proposals | WSDDN · PASCAL VOC 2007 test | mAP: 34.8 -> 39.4 | Helped |
| Use video-trained W-RPN proposals to rank high-recall selective search proposals | OICR · PASCAL VOC 2007 | mAP: 41.2 -> 46.9 | Helped |
| Use video-trained W-RPN proposals to rank high-recall selective search proposals | OICR · PASCAL VOC 2007 | CorLoc: 60.6 -> 66.5 | Helped |
| Use video-trained W-RPN proposals to rank high-recall selective search proposals | OICR · PASCAL VOC 2012 | mAP: 37.9 -> 43.2 | Helped |
| Use video-trained W-RPN proposals to rank high-recall selective search proposals | OICR · PASCAL VOC 2012 | CorLoc: 62.1 -> 67.5 | Helped |
| Use W-RPN proposals learned from videos instead of image-only learned proposals | OICR + LP · PASCAL VOC 2007 | mAP: 45.3 -> 46.9 | Helped |
| Use W-RPN proposals learned from videos instead of image-only learned proposals | OICR + LP · PASCAL VOC 2012 | mAP: 40.8 -> 43.2 | Helped |
| Add W-RPN top-10 proposals alongside edge box proposals, without re-ranking (ablation) | WSDDN · PASCAL VOC 2007 | mAP: +0.6% | Helped |
| Re-rank edge box proposals with the W-RPN ranking loss rather than simply combining proposal sets (ablation) | WSDDN · PASCAL VOC 2007 | mAP | Helped |
| Use high-precision video-derived proposals to rank existing high-recall proposals | Weakly-supervised object detection · PASCAL VOC 2007 | average IoU | Helped |
Closest papers by meaning
- Track and Transfer: Watching Videos to Simulate Strong Human Supervision for Weakly-Supervised Object Detection (CVPR 2016 · similarity 0.90)
- Instance-Aware, Context-Focused, and Memory-Efficient Weakly Supervised Object Detection (CVPR 2020 · similarity 0.90)
- HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection (CVPR 2016 · similarity 0.89)
- Group R-CNN for Weakly Semi-Supervised Object Detection With Points (CVPR 2022 · similarity 0.89)
- Weakly Supervised Object Localization With Progressive Domain Adaptation (CVPR 2016 · similarity 0.89)
- Relational Matching for Weakly Semi-Supervised Oriented Object Detection (CVPR 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.