Research map / Label-efficient and robust learning / Domain adaptation and generalization
Unsupervised Intra-Domain Adaptation for Semantic Segmentation Through Self-Supervision
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 easy-to-hard intra-domain adaptation to AdvEnt | AdvEnt on GTA5 → Cityscapes · GTA5 → Cityscapes val | mIoU: 43.8 -> 46.3 | Helped |
| Add intra-domain adversarial alignment without pseudo-label self-training (ablation) | AdvEnt on GTA5 → Cityscapes · GTA5 → Cityscapes val | mIoU: 43.8 -> 45.1 | Helped |
| Add pseudo-label self-training without intra-domain adversarial alignment (ablation) | AdvEnt on GTA5 → Cityscapes · GTA5 → Cityscapes val | mIoU: 43.8 -> 45.5 | Helped |
| Add entropy normalization by the number of predicted rare classes (ablation) | Ours on GTA5 → Cityscapes · GTA5 → Cityscapes val | mIoU: 46.3 -> 47.0 | Helped |
| Set the easy-split ratio to λ=0.67 rather than λ=0.0 (ablation) | Ours on GTA5 → Cityscapes · GTA5 → Cityscapes val | mIoU: 43.8 -> 46.3 | Helped |
| Set the easy-split ratio to λ=0.67 rather than λ=0.5 (ablation) | Ours on GTA5 → Cityscapes · GTA5 → Cityscapes val | mIoU: 45.2 -> 46.3 | Helped |
| Set the easy-split ratio to λ=0.67 rather than λ=0.6 (ablation) | Ours on GTA5 → Cityscapes · GTA5 → Cityscapes val | mIoU: 46.0 -> 46.3 | Helped |
| Set the easy-split ratio to λ=0.67 rather than λ=0.7 (ablation) | Ours on GTA5 → Cityscapes · GTA5 → Cityscapes val | mIoU: 45.6 -> 46.3 | Helped |
| Set the easy-split ratio to λ=0.67 rather than λ=1.0 (ablation) | Ours on GTA5 → Cityscapes · GTA5 → Cityscapes val | mIoU: 45.5 -> 46.3 | Helped |
| Add entropy-ranked intra-domain adaptation | Semantic segmentation UDA · SYNTHIA → Cityscapes val, 16 classes | mIoU: 40.8 -> 41.7 | Helped |
| Add entropy-ranked intra-domain adaptation | Semantic segmentation UDA · SYNTHIA → Cityscapes val, 13 classes | mIoU: 47.6 -> 48.9 | Helped |
| Add entropy-ranked intra-domain adaptation | AdaptSegNet on Synscapes → Cityscapes · Synscapes → Cityscapes val | mIoU: 52.7 -> 54.2 | Helped |
Closest papers by meaning
- Self-Supervised Augmentation Consistency for Adapting Semantic Segmentation (CVPR 2021 · similarity 0.91)
- Unsupervised Model Adaptation for Continual Semantic Segmentation (AAAI 2021 · similarity 0.91)
- Source-Free Domain Adaptation for Semantic Segmentation (CVPR 2021 · similarity 0.91)
- Balanced Learning for Domain Adaptive Semantic Segmentation (ICML 2025 · similarity 0.90)
- Universal Domain Adaptation for Semantic Segmentation (CVPR 2025 · similarity 0.90)
- SemiDAViL: Semi-supervised Domain Adaptation with Vision-Language Guidance for Semantic Segmentation (CVPR 2025 · similarity 0.89)
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.