List your company
ResearchFormerly WizWand

Semantic Segmentation on Cityscapes 8 classes (VKITTI2 context)

79.63mIoUMTL target
48.294856.429964.56572.7001Jun 17, 2022Jan 2, 2023Jul 21, 2023Feb 6, 2024Aug 24, 2024Mar 12, 2025Sep 28, 2025
Updated 7mo ago

Evaluation results

MethodLinks
MTL targetArchitecture=Multi-Task Learning, Strategy=Oracle (Target Labels)2022.0679.63
MTL targetBackbone=DeepLabV2, UDA setting=False, Learning protocol (STL/MTL)=MTL, Source/Target domain evaluation protocol=Target-supervised (Upper Bound)2025.0979.63
STL targetArchitecture=Single-Task Learning, Strategy=Oracle (Target Labels)2022.0677.1
STL targetBackbone=DeepLabV2, UDA setting=False, Learning protocol (STL/MTL)=STL, Source/Target domain evaluation protocol=Target-supervised (Upper Bound)2025.0977.1
VTAGMLBackbone=Swin, UDA setting=True, Learning protocol (STL/MTL)=MTL2025.0970.93
FAMDABackbone=SegFormer, UDA setting=True, Learning protocol (STL/MTL)=MTL2025.0968.4
MulTBackbone=Swin, UDA setting=True, Learning protocol (STL/MTL)=MTL2025.0966.12
FAMDABackbone=DeepLabV2, UDA setting=True, Learning protocol (STL/MTL)=MTL2025.0965.86
Swin-UDABackbone=Swin, UDA setting=True, Learning protocol (STL/MTL)=STL2025.0963.88
mTEB (Ours)Architecture=Multi-Task Learning with mTEB, Strategy=Unsupervised Domain Adaptation2022.0663.76
XTAMBackbone=DeepLabV2, UDA setting=True, Learning protocol (STL/MTL)=MTL2025.0963.76
STL-UDAArchitecture=Single-Task Learning, Strategy=Unsupervised Domain Adaptation2022.0661.6
STL-UDABackbone=DeepLabV2, UDA setting=True, Learning protocol (STL/MTL)=STL2025.0961.6
STL sourceArchitecture=Single-Task Learning, Strategy=Source only2022.0658.77
STL sourceBackbone=DeepLabV2, UDA setting=False, Learning protocol (STL/MTL)=STL, Source/Target domain evaluation protocol=Source-only (No adaptation)2025.0958.77
MTL-UDAArchitecture=Multi-Task Learning, Strategy=Unsupervised Domain Adaptation2022.0657.26
MTL-UDABackbone=DeepLabV2, UDA setting=True, Learning protocol (STL/MTL)=MTL2025.0957.26
MTL sourceArchitecture=Multi-Task Learning, Strategy=Source only2022.0649.5
MTL sourceBackbone=DeepLabV2, UDA setting=False, Learning protocol (STL/MTL)=MTL, Source/Target domain evaluation protocol=Source-only (No adaptation)2025.0949.5