Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/DAFormer

DAFormer

Reported on 28 benchmarks across 8 tasks · 2 papers · 15 SOTA

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

Medical8 results

Image GenerationonSYNTHIA-to-Cityscapes
MIoU (16 classes)· 2021-11-29
60.9
best: 69.3 (DCF)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Semantic SegmentationonDark Zurich
mIoU· 2021-11-29
53.8
best: 63.9 (Refign (HRDA))
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Semantic SegmentationonGTAV-to-Cityscapes Labels
mIoU· 2021-11-29
68.3
best: 75.9 (MIC)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Semantic SegmentationonSYNTHIA-to-Cityscapes
Mean IoU· 2021-11-29
60.9
best: 68.2 (HRDA + PiPa)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Image GenerationonSYNTHIA-to-Cityscapes
mIoU (13 classes)· 2021-11-29
67.4
best: 74.8 (HRDA + PiPa)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Image GenerationonGTAV-to-Cityscapes Labels
mIoU· 2021-11-29
68.3
best: 77.7 (DCF)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Image GenerationonGTAV-to-Cityscapes Labels
mIoU· 2021-11-29
68.3
best: 77.7 (DCF)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Image GenerationonSYNTHIA-to-Cityscapes
MIoU (13 classes)· 2021-11-29
67.4
best: 75.9 (DCF)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887

Methodology7 results

Domain AdaptationonSYNTHIA-to-Cityscapes
mIoU· 2021-11-29
60.9
best: 78.1 (HALO)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Domain AdaptationonGTA5 to Cityscapes
mIoU· 2021-11-29
68.3
best: 77.8 (HALO)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Domain AdaptationonCityscapes to ACDC
mIoU· 2021-11-29
55.4
best: 78.8 (SoRA)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Domain AdaptationonSYNTHIA-to-Cityscapes
mIoU (13 classes)· 2021-11-29
67.4
best: 75.9 (DCF)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Domain AdaptationonGTA-to-Avg(Cityscapes,BDD,Mapillary)
mIoU· 2023-04-26
51.73
best: 4489 (Self-adaptation (ResNet - 101))
Domain Adaptive and Generalizable Network Architectures and Training Strategies for Semantic Image Segmentation arXiv:2304.13615
Domain AdaptationonGTAV-to-Cityscapes Labels
mIoU· 2021-11-29
68.3
best: 75.9 (MIC)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Domain AdaptationonSYNTHIA-to-Cityscapes
mIoU· 2021-11-29
60.9
best: 78.1 (HALO)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887

Computer Vision6 results

Image-to-Image TranslationonSYNTHIA-to-Cityscapes
MIoU (16 classes)· 2021-11-29
60.9
best: 69.3 (DCF)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Domain GeneralizationonGTA-to-Avg(Cityscapes,BDD,Mapillary)
mIoU· 2023-04-26
51.73
best: 4489 (Self-adaptation (ResNet - 101))
Domain Adaptive and Generalizable Network Architectures and Training Strategies for Semantic Image Segmentation arXiv:2304.13615
Image-to-Image TranslationonSYNTHIA-to-Cityscapes
mIoU (13 classes)· 2021-11-29
67.4
best: 74.8 (HRDA + PiPa)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Image-to-Image TranslationonGTAV-to-Cityscapes Labels
mIoU· 2021-11-29
68.3
best: 77.7 (DCF)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Image-to-Image TranslationonGTAV-to-Cityscapes Labels
mIoU· 2021-11-29
68.3
best: 77.7 (DCF)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Image-to-Image TranslationonSYNTHIA-to-Cityscapes
MIoU (13 classes)· 2021-11-29
67.4
best: 75.9 (DCF)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887

Miscellaneous5 results

1 Image, 2*2 StitchingonSYNTHIA-to-Cityscapes
MIoU (16 classes)· 2021-11-29
60.9
best: 69.3 (DCF)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
1 Image, 2*2 StitchingonSYNTHIA-to-Cityscapes
mIoU (13 classes)· 2021-11-29
67.4
best: 74.8 (HRDA + PiPa)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
1 Image, 2*2 StitchingonGTAV-to-Cityscapes Labels
mIoU· 2021-11-29
68.3
best: 77.7 (DCF)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
1 Image, 2*2 StitchingonGTAV-to-Cityscapes Labels
mIoU· 2021-11-29
68.3
best: 77.7 (DCF)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
1 Image, 2*2 StitchingonSYNTHIA-to-Cityscapes
MIoU (13 classes)· 2021-11-29
67.4
best: 75.9 (DCF)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887

Other3 results

Unsupervised Domain AdaptationonSYNTHIA-to-Cityscapes
mIoU· 2021-11-29
60.9
best: 69.3 (DCF)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Unsupervised Domain AdaptationonSYNTHIA-to-Cityscapes
mIoU (13 classes)· 2021-11-29
67.4
best: 75.9 (DCF)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
Unsupervised Domain AdaptationonGTAV-to-Cityscapes Labels
mIoU· 2021-11-29
68.3
best: 75.9 (MIC)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887

Audio3 results

10-shot image generationonDark Zurich
mIoU· 2021-11-29
53.8
best: 63.9 (Refign (HRDA))
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
10-shot image generationonGTAV-to-Cityscapes Labels
mIoU· 2021-11-29
68.3
best: 75.9 (MIC)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887
10-shot image generationonSYNTHIA-to-Cityscapes
Mean IoU· 2021-11-29
60.9
best: 68.2 (HRDA + PiPa)
SOTA
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation arXiv:2111.14887