Research map / Detection and segmentation / Weakly supervised detection and segmentation
Activation Modulation and Recalibration Scheme for Weakly 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 channel AMM to the single-branch baseline (ablation) | AMR · PASCAL VOC2012 train CAMs | mIoU: 48.3 -> 52.9 | Helped |
| add spatial AMM to the single-branch baseline (ablation) | AMR · PASCAL VOC2012 train CAMs | mIoU: 48.3 -> 53.5 | Helped |
| add both channel and spatial AMM (ablation) | AMR · PASCAL VOC2012 train CAMs | mIoU: 48.3 -> 54.9 | Helped |
| add cross pseudo supervision to the model with both AMM modules (ablation) | AMR · PASCAL VOC2012 train CAMs | mIoU: 54.9 -> 56.8; 1.9% | Helped |
| use threshold modulation instead of Gaussian modulation (ablation) | AMR · PASCAL VOC2012 train CAMs | mIoU: 56.8 -> 50.1 | Hurt |
| set recalibration coefficient ξ to 0.1 instead of 0.5 (ablation) | AMR · PASCAL VOC2012 train CAMs | mIoU: 56.8 -> 49.2 | Hurt |
| set recalibration coefficient ξ to 0.3 instead of 0.5 (ablation) | AMR · PASCAL VOC2012 train CAMs | mIoU: 56.8 -> 53.4 | Hurt |
| set recalibration coefficient ξ to 0.7 instead of 0.5 (ablation) | AMR · PASCAL VOC2012 train CAMs | mIoU: 56.8 -> 54.5 | Hurt |
| set recalibration coefficient ξ to 0.9 instead of 0.5 (ablation) | AMR · PASCAL VOC2012 train CAMs | mIoU: 56.8 -> 50.7 | Hurt |
| incorporate AMR into IRNet | IRNet · PASCAL VOC2012 train CAMs | mIoU: 48.3 -> 56.8; 8.5% | Helped |
| incorporate AMR into SEAM | SEAM · PASCAL VOC2012 train CAMs | mIoU: 55.4 -> 57.9; 2.5% | Helped |
Closest papers by meaning
- Progressive Feature Self-Reinforcement for Weakly Supervised Semantic Segmentation (AAAI 2024 · similarity 0.89)
- From SAM to CAMs: Exploring Segment Anything Model for Weakly Supervised Semantic Segmentation (CVPR 2024 · similarity 0.89)
- Weakly Supervised Semantic Segmentation via Progressive Confidence Region Expansion (CVPR 2025 · similarity 0.89)
- Class Tokens Infusion for Weakly Supervised Semantic Segmentation (CVPR 2024 · similarity 0.89)
- SFC: Shared Feature Calibration in Weakly Supervised Semantic Segmentation (AAAI 2024 · similarity 0.88)
- W2P: Switching from Weak Supervision to Partial Supervision for Semantic Segmentation (AAAI 2024 · similarity 0.88)
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.