Research map / Detection and segmentation / Weakly supervised detection and segmentation
Two-Phase Learning for Weakly Supervised Object Localization
Dahun Kim, Donghyeon Cho, Donggeun Yoo, In So Kweon
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 |
|---|---|---|---|
| replace the SEC localization module with two-phase heat maps | SEC · VOC 2012 segmentation val | mIoU: 50.7 -> 53.1 | Improved |
| replace the SEC localization module with two-phase heat maps | SEC · VOC 2012 segmentation test | mIoU: 51.7 -> 53.8 | Improved |
| apply two-phase learning to CCNN | CCNN · VOC 2012 segmentation val | mIoU: 34.5 -> 35.7 | Improved |
| combine second-phase heat maps with first by weighted map voting (ablation) | two-phase heat maps · VOC 2012 val per-class saliency | AP: 32.5 -> 37.7 | Improved |
| use the second network instead of the first for location prediction | two-phase networks · VOC 2012 main val | location mAP: 88.1 -> 82.6 | Hurt |
| add a third phase with 40% threshold (ablation) | two-phase learning · VOC 2012 main val | location mAP: 82.6 -> 70.5 | Hurt |
| train SEC with seed, expand and constrain losses versus seed and constrain only (ablation) | SEC · VOC 2012 segmentation val | mIoU: +0.3% mIoU | Improved |
Closest papers by meaning
- Weakly Supervised Localization Using Deep Feature Maps (ECCV 2016 · similarity 0.90)
- Weakly-Supervised Semantic Segmentation by Iteratively Mining Common Object Features (CVPR 2018 · similarity 0.90)
- Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi-Supervised Semantic Segmentation (CVPR 2018 · similarity 0.90)
- Incorporating Network Built-in Priors in Weakly-Supervised Semantic Segmentation (TPAMI 2017 · similarity 0.90)
- Soft Proposal Networks for Weakly Supervised Object Localization (ICCV 2017 · similarity 0.89)
- Looking Beyond Single Images for Weakly Supervised Semantic Segmentation Learning (TPAMI 2022 · similarity 0.89)
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.
