Research map / Detection and segmentation / Weakly supervised detection and segmentation
Human-Like Delicate Region Erasing Strategy for Weakly Supervised Detection
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 |
|---|---|---|---|
| Replace hard-threshold post-processing with mean-value threshold (ablation) | Erasing agent · PASCAL VOC 2007 test detection | mAP: 34.9 -> 38.4 | Helped |
| Use dilation CNN heat maps with mean threshold post-processing (ablation) | Erasing agent · PASCAL VOC 2007 test detection | mAP: 38.4 -> 41.2 | Helped |
| Replace hard threshold with mean threshold (ablation) | Erasing agent · PASCAL VOC 2007 trainval | CorLoc: 32.7 -> 34.7 | Helped |
| Replace hard threshold with mean threshold (ablation) | Erasing agent · PASCAL VOC 2007 test pointing localization | accuracy (%): 76.0 -> 76.7 | Helped |
| Count a hit if any of the top-3 proposals locates the object instead of top-1 | Erasing agent · PASCAL VOC 2007 trainval | CorLoc: 34.7 -> 40.6 | Helped |
| Increase total erasing steps T from 3 to 7 (ablation) | Erasing agent · PASCAL VOC 2007 test detection | AP per category | Mixed |
| Lower classification confidence parameter xi to 0.4 (ablation) | Erasing agent reward · PASCAL VOC 2007 test, aeroplane | AP | Helped |
| Raise classification reward sigma to 3.2 (ablation) | Erasing agent reward · PASCAL VOC 2007 test, aeroplane | AP | Helped |
| Set erasing degree reward beta to 0.5 (ablation) | Erasing agent reward · PASCAL VOC 2007 test, aeroplane | AP | Helped |
| Set termination penalty zeta to -0.5 (ablation) | Erasing agent reward · PASCAL VOC 2007 test, aeroplane | AP | Helped |
Closest papers by meaning
- Erasing Integrated Learning: A Simple Yet Effective Approach for Weakly Supervised Object Localization (CVPR 2020 · similarity 0.89)
- Deep Self-Taught Learning for Weakly Supervised Object Localization (CVPR 2017 · similarity 0.88)
- Tell Me What They're Holding: Weakly-Supervised Object Detection with Transferable Knowledge from Human-Object Interaction (AAAI 2020 · similarity 0.87)
- Bridging the Gap Between Classification and Localization for Weakly Supervised Object Localization (CVPR 2022 · similarity 0.86)
- Weakly Supervised Deep Detection Networks (CVPR 2016 · similarity 0.86)
- You Reap What You Sow: Using Videos to Generate High Precision Object Proposals for Weakly-Supervised Object Detection (CVPR 2019 · similarity 0.85)
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.