Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain-Agnostic Prior for Transfer Semantic Segmentation
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 word2vec domain-agnostic prior to DACS | DACS · GTAv→Cityscapes | mIOU: 52.1 -> 55.0; 2.9% gain | Helped |
| Add word2vec domain-agnostic prior to DACS | DACS · SYNTHIA→Cityscapes | mIOU: 48.9 -> 50.2; 1.3% gain | Helped |
| Use DAP output as ProDA's first-stage pseudo labels | ProDA · GTAv→Cityscapes | mIOU: +2.3% | Helped |
| Use DAP output as ProDA's first-stage pseudo labels | ProDA · SYNTHIA→Cityscapes | mIOU: +4.3% | Helped |
| Use random class vectors as the prior (ablation) | DACS · GTAv→Cityscapes | mIOU: 52.1 -> 52.3; 0.2% gain | Helped |
| Use random class vectors as the prior (ablation) | DACS · SYNTHIA→Cityscapes | mIOU: 48.9 -> 48.0; 0.9% accuracy drop | Hurt |
| Use one-hot class vectors as the prior (ablation) | DACS · GTAv→Cityscapes | mIOU: 52.1 -> 53.0; 0.9% gain | Helped |
| Use one-hot class vectors as the prior (ablation) | DACS · SYNTHIA→Cityscapes | mIOU: 48.9 -> 49.8; 0.9% gain | Helped |
| Use CLIP language embeddings as the prior (ablation) | DACS · GTAv→Cityscapes | mIOU: 52.1 -> 54.6; 2.5% gain | Helped |
| Use CLIP language embeddings as the prior (ablation) | DACS · SYNTHIA→Cityscapes | mIOU: 48.9 -> 50.6; 1.7% gain | Helped |
| Remove feature interpolation when down-sampling the embedding map (ablation) | DACS+DAP · GTAv→Cityscapes | mIOU: 53.8 -> 55.0 | Hurt |
| Remove feature interpolation when down-sampling the embedding map (ablation) | DACS+DAP · SYNTHIA→Cityscapes | mIOU: 49.6 -> 50.2 | Hurt |
Closest papers by meaning
- Unsupervised Model Adaptation for Continual Semantic Segmentation (AAAI 2021 · similarity 0.91)
- Universal Domain Adaptation for Semantic Segmentation (CVPR 2025 · similarity 0.90)
- Open-Set Domain Adaptation for Semantic Segmentation (CVPR 2024 · similarity 0.89)
- Towards Unsupervised Domain Bridging via Image Degradation in Semantic Segmentation (NeurIPS 2025 · similarity 0.89)
- UniDSeg: Unified Cross-Domain 3D Semantic Segmentation via Visual Foundation Models Prior (NeurIPS 2024 · similarity 0.88)
- Balanced Learning for Domain Adaptive Semantic Segmentation (ICML 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.