Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly Supervised Semantic Segmentation Using Superpixel Pooling Network
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 |
|---|---|---|---|
| superpixel pooling (SP-CAM) instead of CAM for annotation generation (ablation) | initial annotation generation · PASCAL VOC 2012 train | annotation accuracy: 30.5% -> 43.8% | Helped |
| train DecoupledNet on SPN annotations (ablation) | SPN · PASCAL VOC 2012 val | mean IoU: 40.0 % -> 48.6 % | Helped |
| add a second round of iterative training (ablation) | DecoupledNet · PASCAL VOC 2012 val | mean IoU: 48.6 % -> 50.2 % | Helped |
| annotation sanitization with image-level labels (ablation) | annotation generation · PASCAL VOC 2012 train | annotation accuracy | Helped |
| train on 300 most reliable annotations per class instead of all (ablation) | DecoupledNet · PASCAL VOC 2012 val | mean IoU | Helped |
| full method vs MIL+Seg | weakly supervised segmentation · PASCAL VOC 2012 test | mean IoU: 40.6 % -> 46.9 % | Helped |
Closest papers by meaning
- Semantic-Aware Superpixel for Weakly Supervised Semantic Segmentation (AAAI 2023 · similarity 0.93)
- Background-Aware Pooling and Noise-Aware Loss for Weakly-Supervised Semantic Segmentation (CVPR 2021 · similarity 0.90)
- Combining Bottom-Up, Top-Down, and Smoothness Cues for Weakly Supervised Image Segmentation (CVPR 2017 · similarity 0.89)
- Semantic Projection Network for Zero- and Few-Label Semantic Segmentation (CVPR 2019 · similarity 0.89)
- Reliability Does Matter: An End-to-End Weakly Supervised Semantic Segmentation Approach (AAAI 2020 · similarity 0.89)
- W2P: Switching from Weak Supervision to Partial Supervision for Semantic Segmentation (AAAI 2024 · similarity 0.86)
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.