Research map / Detection and segmentation / Weakly supervised detection and segmentation
L2G: A Simple Local-to-Global Knowledge Transfer Framework for Weakly Supervised Semantic Segmentation
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 |
|---|---|---|---|
| Transfer local-patch attention maps to the global network | CAM-based weakly supervised segmentation · PASCAL VOC 2012 val | mIoU: 47.5 -> 54.9 | Helped |
| Use sliding-window inference instead of local-to-global transfer (ablation) | CAM-based weakly supervised segmentation · PASCAL VOC 2012 val | mIoU: 47.5 -> 47.2 | Hurt |
| Add saliency-based shape transfer to local-to-global attention transfer (ablation) | L2G · PASCAL VOC 2012 val | mIoU: 54.9 -> 72.1 | Helped |
| Use local-to-global transfer rather than local patches alone (ablation) | Local-view classification network · PASCAL VOC 2012 val | mIoU: 50.0 -> 54.9 | Helped |
| Switch local-view sampling from random to uniform (ablation) | L2G · PASCAL VOC 2012 train | pseudo-label mIoU: 68.8% v.s. 68.5% | No clear effect |
| Increase the number of local views (ablation) | L2G · PASCAL VOC 2012 | pseudo-label mIoU | Helped |
| Increase patch size beyond 320×320 (ablation) | L2G · PASCAL VOC 2012 | pseudo-label mIoU | Hurt |
| Add classification loss to the global network (ablation) | L2G · PASCAL VOC 2012 trainaug | pseudo-label mIoU: 70.3% to 53.8% | Hurt |
| Use separate rather than shared local and global backbones (ablation) | L2G · PASCAL VOC 2012 val | mIoU: +1.2% | Helped |
| Use L2G rather than EPS | Weakly supervised segmentation · MS COCO 2014 val | mIoU: 35.7 -> 44.2 | Helped |
Closest papers by meaning
- Embedded Discriminative Attention Mechanism for Weakly Supervised Semantic Segmentation (CVPR 2021 · similarity 0.90)
- Tell Me Where to Look: Guided Attention Inference Network (CVPR 2018 · similarity 0.89)
- Partial Class Activation Attention for Semantic Segmentation (CVPR 2022 · similarity 0.89)
- MoRe: Class Patch Attention Needs Regularization for Weakly Supervised Semantic Segmentation (AAAI 2025 · similarity 0.87)
- W2P: Switching from Weak Supervision to Partial Supervision for Semantic Segmentation (AAAI 2024 · similarity 0.87)
- Separate and Conquer: Decoupling Co-occurrence via Decomposition and Representation for Weakly Supervised Semantic Segmentation (CVPR 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.