DIFF

Reported on 4 benchmarks across 2 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology2 results

Domain AdaptationonGTA-to-Avg(Cityscapes,BDD,Mapillary)
mIoU· 2024-06-02
57.15
best: 4489 (Self-adaptation (ResNet - 101))
Diffusion Features to Bridge Domain Gap for Semantic Segmentation arXiv:2406.00777
Domain AdaptationonGTA5-to-Cityscapes
mIoU· 2024-06-02
58.01
best: 74.4 (CLUDA+HRDA)
Diffusion Features to Bridge Domain Gap for Semantic Segmentation arXiv:2406.00777

Computer Vision2 results

Domain GeneralizationonGTA-to-Avg(Cityscapes,BDD,Mapillary)
mIoU· 2024-06-02
57.15
best: 4489 (Self-adaptation (ResNet - 101))
Diffusion Features to Bridge Domain Gap for Semantic Segmentation arXiv:2406.00777
Domain GeneralizationonGTA5-to-Cityscapes
mIoU· 2024-06-02
58.01
best: 68.88 (tqdm (EVA02-CLIP-L))
Diffusion Features to Bridge Domain Gap for Semantic Segmentation arXiv:2406.00777