Research map / Label-efficient and robust learning / Semi-supervised learning
Semi-Supervised Semantic Segmentation With Pixel-Level Contrastive Learning From a Class-Wise Memory Bank
Iñigo Alonso, Alberto Sabater, David Ferstl, Luis Montesano, Ana C. Murillo
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 contrastive loss to supervised loss alone (ablation) | Teacher-student segmentation · Cityscapes val (1/30 labels, ablation) | mIoU: 49.5 -> 54.4 | Improved |
| add pseudo-label loss (ablation) | Teacher-student segmentation · Cityscapes val (1/30 labels, ablation) | mIoU: 49.5 -> 56.7 | Improved |
| add entropy minimization loss to supervised loss (ablation) | Teacher-student segmentation · Cityscapes val (1/30 labels, ablation) | mIoU: 49.5 -> 52.2 | Improved |
| add contrastive loss to sup + pseudo + ent (ablation) | Full method · Cityscapes val (1/30 labels, ablation) | mIoU: 57.4 -> 59.4 | Improved |
| add entropy loss to sup + pseudo + contr (ablation) | Full method · Cityscapes val (1/30 labels, ablation) | mIoU: 59.0 -> 59.4 | Improved |
| add pseudo-label loss to sup + ent + contr (ablation) | Full method · Cityscapes val (1/30 labels, ablation) | mIoU: 57.3 -> 59.4 | Improved |
| set contrastive weight to 0.1 instead of 10^4 (ablation) | Contrastive loss · Cityscapes val (1/30 labels, ablation) | mIoU: 50.3 -> 59.4 | Improved |
| set contrastive weight to 10^-4 instead of 0.1 (ablation) | Contrastive loss · Cityscapes val (1/30 labels, ablation) | mIoU: 59.4 -> 57.6 | Hurt |
| increase memory bank size per class from 32 to 256 (ablation) | Memory bank · Cityscapes val (1/30 labels, ablation) | mIoU: 58.7 -> 59.4 | Improved |
| increase memory bank size from 256 to 512 (ablation) | Memory bank · Cityscapes val (1/30 labels, ablation) | mIoU: 59.4 -> 59.3 | No clear effect |
| add Feature Quality Filter to base contrastive module (ablation) | Contrastive module · Cityscapes val (1/30 labels, ablation) | mIoU: 58.3 -> 59.0 | Improved |
| extract memory features with teacher (ablation) | Contrastive module · Cityscapes val (1/30 labels, ablation) | mIoU: 58.3 -> 58.7 | Improved |
Closest papers by meaning
- Semi-Supervised Semantic Segmentation With High- and Low-Level Consistency (TPAMI 2019 · similarity 0.92)
- Pixel Contrastive-Consistent Semi-Supervised Semantic Segmentation (ICCV 2021 · similarity 0.92)
- Boosting Semi-Supervised Semantic Segmentation with Probabilistic Representations (AAAI 2023 · similarity 0.92)
- SemiDAViL: Semi-supervised Domain Adaptation with Vision-Language Guidance for Semantic Segmentation (CVPR 2025 · similarity 0.91)
- Semi-Supervised Semantic Segmentation With Directional Context-Aware Consistency (CVPR 2021 · similarity 0.91)
- Semi-Supervised Semantic Segmentation under Label Noise via Diverse Learning Groups (ICCV 2023 · similarity 0.91)
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.
