Research map / Detection and segmentation / Weakly supervised detection and segmentation
Bootstrapping the Performance of Webly Supervised Semantic Segmentation
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 |
|---|---|---|---|
| add web-domain training with Web-SEC to the target-domain Initial-SEC (ablation) | Bootstrapping webly supervised segmentation · PASCAL VOC 2012 val | IoU: 49.3 -> 52.6 | Helped |
| add Grabcut refinement to the web-domain masks (ablation) | Bootstrapping webly supervised segmentation · PASCAL VOC 2012 val | IoU: 52.6 -> 55.7 | Helped |
| train Web-FCN on the refined web masks (ablation) | Bootstrapping webly supervised segmentation · PASCAL VOC 2012 val | IoU: 55.7 -> 56.6 | Helped |
| add post-processing with multi-scale inference and dense-CRF (ablation) | Bootstrapping webly supervised segmentation · PASCAL VOC 2012 val | IoU: 56.6 -> 58.8 | Helped |
| reduce the filtered web image set from 76.7k to 58.1k (ablation) | Bootstrapping webly supervised segmentation · PASCAL VOC 2012 val | IoU: 56.6 -> 56.4 | No clear effect |
| reduce the filtered web image set to 39.1k (ablation) | Bootstrapping webly supervised segmentation · PASCAL VOC 2012 val | IoU: 56.6 -> 56.3 | No clear effect |
| reduce the filtered web image set to 20.0k (ablation) | Bootstrapping webly supervised segmentation · PASCAL VOC 2012 val | IoU: 56.6 -> 56.4 | No clear effect |
| reduce the filtered web image set to 10.0k (ablation) | Bootstrapping webly supervised segmentation · PASCAL VOC 2012 val | IoU: 56.6 -> 56.4 | No clear effect |
| reduce the filtered web image set to 6k (ablation) | Bootstrapping webly supervised segmentation · PASCAL VOC 2012 val | IoU: 56.6 -> 55.7 | No clear effect |
| reduce the filtered web image set to 2k (ablation) | Bootstrapping webly supervised segmentation · PASCAL VOC 2012 val | IoU: 56.6 -> 55.3 | Hurt |
| remove web-image filtering (ablation) | Bootstrapping webly supervised segmentation · PASCAL VOC 2012 val | IoU: 56.6 -> 49.8 | Hurt |
Closest papers by meaning
- Reliability Does Matter: An End-to-End Weakly Supervised Semantic Segmentation Approach (AAAI 2020 · similarity 0.93)
- Progressive Feature Self-Reinforcement for Weakly Supervised Semantic Segmentation (AAAI 2024 · similarity 0.89)
- W2P: Switching from Weak Supervision to Partial Supervision for Semantic Segmentation (AAAI 2024 · similarity 0.89)
- DuPL: Dual Student with Trustworthy Progressive Learning for Robust Weakly Supervised Semantic Segmentation (CVPR 2024 · similarity 0.89)
- Label-Efficient Few-Shot Semantic Segmentation with Unsupervised Meta-Training (AAAI 2024 · similarity 0.89)
- From SAM to CAMs: Exploring Segment Anything Model 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.