Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly-Supervised Semantic Segmentation via Sub-Category Exploration
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 |
|---|---|---|---|
| Add sub-category exploration to CAM | CAM initial response generation · PASCAL VOC 2012 validation set | mIoU: 46.8 -> 49.6 | Helped |
| Add sub-category exploration before random-walk refinement | CAM+RW · PASCAL VOC validation set | mIoU: 57.0 -> 61.2 | Helped |
| Add sub-category exploration | Weakly-supervised segmentation · PASCAL VOC 2012 val | mIoU: 64.9 -> 66.1 | Helped |
| Add sub-category exploration | Weakly-supervised segmentation · PASCAL VOC 2012 test | mIoU: 65.3 -> 65.9 | Helped |
| Add sub-category objective and iterative training (ablation) | Initial CAM response · PASCAL VOC 2012 val | mIoU: 46.8 -> 49.6 | Helped |
| Continue iterative clustering and sub-category training from round #1 to round #3 (ablation) | Initial CAM response · PASCAL VOC 2012 val | mIoU: 48.0 -> 49.6 | Helped |
| Continue iterative clustering and sub-category training from round #1 to round #3 (ablation) | Initial CAM response · PASCAL VOC 2012 val | F-Score: 65.6 -> 67.0 | Helped |
| Use multiple sub-categories (K>1) instead of the original CAM setting (K=1) (ablation) | Sub-category discovery for CAM · PASCAL VOC validation set | mIoU | Helped |
| Vary the sub-category count among K = {5, 10, 20, 50} (ablation) | Sub-category discovery for CAM · PASCAL VOC validation set | mIoU | No clear effect |
Closest papers by meaning
- Weakly-Supervised Semantic Segmentation by Iteratively Mining Common Object Features (CVPR 2018 · similarity 0.91)
- Combining Bottom-Up, Top-Down, and Smoothness Cues for Weakly Supervised Image Segmentation (CVPR 2017 · similarity 0.91)
- Reliability Does Matter: An End-to-End Weakly Supervised Semantic Segmentation Approach (AAAI 2020 · similarity 0.91)
- Universal Weakly Supervised Segmentation by Pixel-to-Segment Contrastive Learning (ICLR 2021 · similarity 0.90)
- Progressive Feature Self-Reinforcement for Weakly Supervised Semantic Segmentation (AAAI 2024 · similarity 0.90)
- Separate and Conquer: Decoupling Co-occurrence via Decomposition and Representation for Weakly Supervised Semantic Segmentation (CVPR 2024 · 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.