Research map / Label-efficient and robust learning / Domain adaptation and generalization
Coarse-To-Fine Domain Adaptive Semantic Segmentation With Photometric Alignment and Category-Center Regularization
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 |
|---|---|---|---|
| Add the complete coarse-to-fine pipeline | DeepLab V3+(Resnet101) on GTA5→Cityscapes · GTA5→Cityscapes | mIoU: 37.6 -> 56.1 | Helped |
| Add global photometric alignment (GPA) (ablation) | Coarse-to-fine segmentation on GTA5→Cityscapes · GTA5→Cityscapes | mIoU: 47.5 -> 56.1 | Helped |
| Add category-oriented triplet loss (CTL) (ablation) | GPA-based segmentation on GTA5→Cityscapes · GTA5→Cityscapes | mIoU: 53.2 -> 56.1 | Helped |
| Add target consistency regularization (TCR) (ablation) | GPA-based segmentation on GTA5→Cityscapes · GTA5→Cityscapes | mIoU: 53.1 -> 56.1 | Helped |
| Add both CTL and TCR (ablation) | GPA-based segmentation on GTA5→Cityscapes · GTA5→Cityscapes | mIoU: 47.3 -> 56.1 | Helped |
| Use GPA rather than frequency alignment (ablation) | Coarse-to-fine segmentation on GTA5→Cityscapes · GTA5→Cityscapes | mIoU: 52.0 -> 56.1 | Helped |
| Use GPA rather than BDL-GAN (ablation) | Coarse-to-fine segmentation on GTA5→Cityscapes · GTA5→Cityscapes | mIoU: 54.5 -> 56.1 | Helped |
| Use the hybrid GPA scheme instead of gamma correction or histogram matching on all Lab channels (ablation) | Coarse alignment on GTA5→Cityscapes · GTA5→Cityscapes | mIoU: 44.5 -> 47.3; 43.3 -> 47.3 | Helped |
| Exclude target-domain pseudo-labels from CTL (ablation) | Category-oriented triplet loss on GTA5→Cityscapes · GTA5→Cityscapes | mIoU: 53.3 -> 56.1 | Helped |
| Use the complete coarse-to-fine pipeline rather than FDA | Semantic segmentation on GTA5→Cityscapes · GTA5→Cityscapes | mIoU: 50.4 -> 56.1 | Helped |
| Use the complete coarse-to-fine pipeline rather than CAG (16 classes) | Semantic segmentation on Synthia→Cityscapes · Synthia→Cityscapes (16 classes) | mIoU: 44.5 -> 48.2 | Helped |
| Use the complete coarse-to-fine pipeline rather than CAG (13 classes) | Semantic segmentation on Synthia→Cityscapes · Synthia→Cityscapes (13 classes) | mIoU*: 52.6 -> 55.5 | Helped |
Closest papers by meaning
- Universal Domain Adaptation for Semantic Segmentation (CVPR 2025 · similarity 0.92)
- Category Anchor-Guided Unsupervised Domain Adaptation for Semantic Segmentation (NeurIPS 2019 · similarity 0.92)
- Unsupervised Model Adaptation for Continual Semantic Segmentation (AAAI 2021 · similarity 0.91)
- Balanced Learning for Domain Adaptive Semantic Segmentation (ICML 2025 · similarity 0.91)
- Open-Set Domain Adaptation for Semantic Segmentation (CVPR 2024 · similarity 0.90)
- Towards Unsupervised Domain Bridging via Image Degradation in Semantic Segmentation (NeurIPS 2025 · similarity 0.90)
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.