Research map / Label-efficient and robust learning / Domain adaptation and generalization
DIRL: Domain-Invariant Representation Learning for Generalizable 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 |
|---|---|---|---|
| Use DIRL instead of ISW | ResNet50 semantic segmentation · GTAV→{Cityscapes, BDD, Mapillary} | mean mIoU: 37.37 -> 40.60 | Helped |
| Use DIRL instead of ISW | ResNet50 semantic segmentation · Cityscapes→{BDD, Synthia, GTAV} | mean mIoU: 40.64 -> 41.61 | Helped |
| Add PGAM without feature-sensitivity guidance (ablation) | DeepLabV3+ ResNet-50 · GTAV→{Cityscapes, BDD, Mapillary} | mIoU: 28.95 -> 30.81; 25.14 -> 26.32; 28.18 -> 29.05 | Helped |
| Directly use feature sensitivity to recalibrate features (ablation) | DeepLabV3+ ResNet-50 · GTAV→{Cityscapes, BDD, Mapillary} | mIoU: 28.95 -> 33.57; 25.14 -> 28.74; 28.18 -> 30.24 | Helped |
| Add Sensitivity Guidance loss (ablation) | DIRL feature recalibration · GTAV→{Cityscapes, BDD, Mapillary} | mIoU: 36.60; 30.66; 33.55 | Helped |
| Use Guided Feature Whiting rather than Instance Selective Whitening (ablation) | DIRL · GTAV→{Cityscapes, BDD, Mapillary} | mIoU: 40.20 -> 41.04; 38.10 -> 39.15; 40.79 -> 41.60 | Helped |
| Set selective ratio α to 0.2 rather than the reported best α=0.3 (ablation) | DIRL · GTAV→{Cityscapes, BDD, Mapillary} | mean mIoU: 38.30 vs 40.60 | Hurt |
Closest papers by meaning
- FisherTune: Fisher-Guided Robust Tuning of Vision Foundation Models for Domain Generalized Segmentation (CVPR 2025 · similarity 0.89)
- Stronger Fewer & Superior: Harnessing Vision Foundation Models for Domain Generalized Semantic Segmentation (CVPR 2024 · similarity 0.88)
- Leveraging Depth and Language for Open-Vocabulary Domain-Generalized Semantic Segmentation (NeurIPS 2025 · similarity 0.88)
- A Simple Recipe for Language-guided Domain Generalized Segmentation (CVPR 2024 · similarity 0.88)
- SoMA: Singular Value Decomposed Minor Components Adaptation for Domain Generalizable Representation Learning (CVPR 2025 · similarity 0.88)
- Exploring Semantic Consistency and Style Diversity for Domain Generalized Semantic Segmentation (AAAI 2025 · 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.