Research map / Detection and segmentation / Weakly supervised detection and segmentation
Self-Regulation for Semantic Segmentation
Dong Zhang, Hanwang Zhang, Jinhui Tang, Xian-Sheng Hua, Qianru Sun
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 MEA loss (multi-exit per-layer classification and segmentation losses) (ablation) | CONTA+SPGNet · PASCAL VOC 2012 val | mIoU (%): 67.1 -> 67.7 | Improved |
| add MEA loss (ablation) | CONTA+SPGNet · MS-COCO 2014 val | mIoU (%): 33.6 -> 34.0 | Improved |
| add MEA loss (ablation) | OCRNet · Cityscapes val | mIoU (%): 80.7 -> 81.1 | Improved |
| add SR-L to MEA (ablation) | CONTA+SPGNet · PASCAL VOC 2012 val | mIoU (%): 67.7 -> 68.1 | Improved |
| add SR-F to MEA (ablation) | CONTA+SPGNet · PASCAL VOC 2012 val | mIoU (%): 67.7 -> 68.2 | Improved |
| add both SR-F and SR-L to MEA (ablation) | CONTA+SPGNet · PASCAL VOC 2012 val | mIoU (%): 67.7 -> 68.5 | Improved |
| add both SR-F and SR-L to MEA (ablation) | CONTA+SPGNet · MS-COCO 2014 val | mIoU (%): 34.0 -> 34.5 | Improved |
| add both SR-F and SR-L to MEA (ablation) | OCRNet · Cityscapes val | mIoU (%): 81.1 -> 82.1 | Improved |
| add SR-L to MEA (ablation) | OCRNet · Cityscapes val | mIoU (%): 81.1 -> 81.6 | Improved |
| add SR-F to MEA (ablation) | OCRNet · Cityscapes val | mIoU (%): 81.1 -> 81.7 | Improved |
| add SR loss to IRNet pseudo masks with DeepLab-v2 | IRNet+DeepLab-v2 · PASCAL VOC 2012 val | mIoU (%): 63.5 -> 64.6 | Improved |
| add SR loss to SEAM with DeepLab-v2 | SEAM+DeepLab-v2 · PASCAL VOC 2012 val | mIoU (%): 64.5 -> 65.6 | Improved |
Closest papers by meaning
- Improving Semantic Segmentation via Efficient Self-Training (TPAMI 2021 · similarity 0.89)
- Single-Stage Semantic Segmentation From Image Labels (CVPR 2020 · similarity 0.88)
- Rethinking Self-Supervised Semantic Segmentation: Achieving End-to-End Segmentation (TPAMI 2024 · similarity 0.87)
- Regularized Loss for Weakly Supervised Single Class Semantic Segmentation (ECCV 2020 · similarity 0.87)
- Learning from Future: A Novel Self-Training Framework for Semantic Segmentation (NeurIPS 2022 · similarity 0.87)
- Progressive Feature Self-Reinforcement for Weakly Supervised Semantic Segmentation (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.
