Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly-Supervised Semantic Segmentation Network With Deep Seeded Region Growing
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 |
|---|---|---|---|
| use DSRG instead of static-label weak supervision | weakly supervised semantic segmentation on PASCAL VOC 2012 · PASCAL VOC 2012 val | mIoU: 58.6 -> 61.4 | Helped |
| use DSRG instead of static-label weak supervision | weakly supervised semantic segmentation on PASCAL VOC 2012 · PASCAL VOC 2012 test | mIoU: 59.2 -> 63.2 | Helped |
| use DSRG instead of the strongest listed baseline | weakly supervised semantic segmentation on COCO · COCO val | mean IoU: 22.4 -> 26.0 | Helped |
| replace the original seeding loss with balanced seeding loss (ablation) | VGG-16 segmentation baseline · PASCAL VOC 2012 val | mIoU: 52.5 -> 53.6 | Helped |
| add DSRG training to the balanced-seeding-loss model (ablation) | VGG-16 segmentation model · PASCAL VOC 2012 val | mIoU: 53.6 -> 57.6 | Helped |
| retrain with predicted segmentation masks as supervision (ablation) | VGG-16 DSRG model · PASCAL VOC 2012 val | mIoU: 57.6 -> 59.0 | Helped |
| perform region growing only for foreground rather than for both foreground and background (ablation) | VGG-16 DSRG model · PASCAL VOC 2012 val | mIoU: 55.9 | Hurt |
| perform region growing only for background rather than for both foreground and background (ablation) | VGG-16 DSRG model · PASCAL VOC 2012 val | mIoU: 54.3 | Hurt |
| vary foreground and background region-growing thresholds (ablation) | VGG-16 DSRG model · PASCAL VOC 2012 val | mIoU: 57.45; 57.59; 57.63; 57.69; 57.66 | No clear effect |
Closest papers by meaning
- Learning Pixel-Level Semantic Affinity With Image-Level Supervision for Weakly Supervised Semantic Segmentation (CVPR 2018 · similarity 0.91)
- Combining Bottom-Up, Top-Down, and Smoothness Cues for Weakly Supervised Image Segmentation (CVPR 2017 · similarity 0.90)
- Webly Supervised Semantic Segmentation (CVPR 2017 · similarity 0.90)
- Reliability Does Matter: An End-to-End Weakly Supervised Semantic Segmentation Approach (AAAI 2020 · similarity 0.90)
- Weakly-Supervised Semantic Segmentation via Sub-Category Exploration (CVPR 2020 · similarity 0.89)
- 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.