Research map / Detection and segmentation / Weakly supervised detection and segmentation
Self Correspondence Distillation for End-to-End 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 VARM to the transformer baseline (ablation) | TSCD framework · PASCAL VOC 2012 val | mIoU: 47.2 -> 56.8 | Helped |
| Add SCD to the VARM variant (ablation) | TSCD framework · PASCAL VOC 2012 val | mIoU: 56.8 -> 63.5 | Helped |
| Add the auxiliary loss to VARM and SCD (ablation) | TSCD framework · PASCAL VOC 2012 val | mIoU: 63.5 -> 64.2 | Helped |
| Add equivariant regularization loss to VARM, SCD, and auxiliary loss (ablation) | TSCD framework · PASCAL VOC 2012 val | mIoU: 64.2 -> 65.0 | Helped |
| Add SCD (ablation) | TSCD with MiT-B1 · PASCAL VOC 2012 val | mIoU: 63.9 -> 65.0 | Helped |
| Add SCD (ablation) | R101-Base · PASCAL VOC 2012 val | mIoU: 53.5 -> 56.5 | Helped |
| Add SCD (ablation) | WR38-Base · PASCAL VOC 2012 val | mIoU: 56.2 -> 59.1 | Helped |
| Add CRF to TSCD | TSCD · PASCAL VOC 2012 val | mIoU: 65.0 -> 67.3 | Helped |
| Add CRF to TSCD | TSCD · MS COCO 2014 val | mIoU: 39.2 -> 40.1 | Helped |
| Use TSCD instead of AFA with the same MiT-B1 backbone | End-to-end WSSS · PASCAL VOC 2012 val | mIoU: 63.8 -> 65.0 | Helped |
| Use TSCD instead of AFA with the same MiT-B1 backbone | End-to-end WSSS · MS COCO 2014 val | mIoU: 38.0 -> 39.2 | Helped |
Closest papers by meaning
- Progressive Feature Self-Reinforcement for Weakly Supervised Semantic Segmentation (AAAI 2024 · similarity 0.91)
- Learning Affinity From Attention: End-to-End Weakly-Supervised Semantic Segmentation With Transformers (CVPR 2022 · similarity 0.91)
- From SAM to CAMs: Exploring Segment Anything Model for Weakly Supervised Semantic Segmentation (CVPR 2024 · similarity 0.88)
- W2P: Switching from Weak Supervision to Partial Supervision for Semantic Segmentation (AAAI 2024 · similarity 0.88)
- $\mathbf{D^{3}S^{2}}$: Diffusion-Guided Dataset Distillation for Semantic Segmentation (NeurIPS 2026 · similarity 0.87)
- DuPL: Dual Student with Trustworthy Progressive Learning for Robust 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.