Semantic Segmentation on Cityscapes GTA5 to Cityscapes adaptation (val)
77.4mIoU (Overall)DFUDA
Evaluation Results
| Method | Links | |||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DFUDASource Domain=GTA2026.05 | 77.4 | 98.1 | 83.1 | 93.5 | 64.1 | 60.4 | 61.2 | 68.3 | 73.3 | 91.4 | 52.1 | 95 | 80.3 | 55.2 | 94.8 | 87.7 | 91.8 | 83.4 | 66.1 | 70 | — | |
| COPTSource Domain=GTA2026.05 | 76.1 | 97.6 | 80.9 | 91.6 | 62.1 | 55.9 | 59.3 | 66.7 | 70.5 | 91.9 | 53 | 94.4 | 80 | 55.6 | 94.7 | 87.1 | 88.6 | 82.1 | 65 | 68.8 | — | |
| MICSource Domain=GTA2026.05 | 75.9 | 97.4 | 80.1 | 91.7 | 61.2 | 56.9 | 59.7 | 66 | 71.3 | 91.7 | 51.4 | 94.3 | 79.8 | 56.1 | 94.6 | 85.4 | 90.3 | 80.4 | 64.5 | 68.5 | — | |
| HRDASource Domain=GTA2026.05 | 73.8 | 96.4 | 74.4 | 91 | 61.6 | 51.5 | 57.1 | 63.9 | 69.3 | 91.3 | 48.4 | 94.2 | 79 | 52.9 | 93.9 | 84.1 | 85.7 | 75.9 | 63.9 | 67.5 | — | |
| SePiCo (DistCL)Backbone=MiT-B5, Source Domain=GTAV, Adaptation Protocol=Distributed Contrastive Learning2022.04 | 70.3 | 96.9 | 76.7 | 89.7 | 55.5 | 49.5 | 53.2 | 60 | 64.5 | 90.2 | 50.3 | 90.8 | 74.5 | 44.2 | 93.3 | 77 | 79.5 | 63.6 | 61 | 65.3 | — | |
| SePiCo (BankCL)Backbone=MiT-B5, Source Domain=GTAV, Adaptation Protocol=Bank-based Contrastive Learning2022.04 | 68.7 | 96.3 | 73.6 | 89.6 | 53.7 | 47.8 | 53.8 | 60.8 | 60 | 89.9 | 48.8 | 91.5 | 74.6 | 45.1 | 93.1 | 74.8 | 73.8 | 51.5 | 60.3 | 65.3 | — | |
| SePiCo (ProtoCL)Backbone=MiT-B5, Source Domain=GTAV, Adaptation Protocol=Prototypical Contrastive Learning2022.04 | 68.5 | 96.1 | 72.9 | 89.7 | 54.4 | 48.8 | 53.5 | 60.4 | 65.3 | 90 | 48.4 | 91.6 | 75.2 | 47.1 | 93.3 | 74.4 | 74.6 | 41.2 | 58.8 | 65.9 | — | |
| DAFormerBackbone=MiT-B5, Source Domain=GTAV2022.04 | 68.3 | 95.7 | 70.2 | 89.4 | 53.5 | 48.1 | 49.6 | 55.8 | 59.4 | 89.9 | 47.9 | 92.5 | 72.2 | 44.7 | 92.3 | 74.5 | 78.2 | 65.1 | 55.9 | 61.8 | — | |
| DAFormerSource Domain=GTA2026.05 | 68.3 | 95.7 | 70.2 | 89.4 | 53.5 | 48.1 | 49.6 | 55.8 | 59.4 | 89.9 | 47.9 | 92.5 | 72.2 | 44.7 | 92.3 | 74.5 | 78.2 | 65.1 | 55.9 | 61.8 | — | |
| TransDA-BBackbone=Swin-B ViT, Source Domain=GTAV2022.04 | 63.9 | 94.7 | 64.2 | 89.2 | 48.1 | 45.8 | 50.1 | 60.2 | 40.8 | 90.4 | 50.2 | 93.7 | 76.7 | 47.6 | 92.5 | 56.8 | 60.1 | 47.6 | 49.6 | 55.4 | — | |
| ProDASource Domain=GTA2026.05 | 57.5 | 87.8 | 56 | 79.7 | 46.3 | 44.8 | 45.6 | 53.5 | 53.5 | 88.6 | 45.2 | 82.1 | 70.7 | 39.2 | 88.8 | 45.5 | 59.4 | 1 | 48.9 | 56.4 | — | |
| TestMate + DT-STAdaptation Setting=SFDA, Backpropagation (BP)=true, Backbone=ResNet-101, Architecture=DeepLabV22026.07 | 54.9 | 87.8 | 46 | 83.7 | 26.5 | 24.9 | 39.4 | 47 | 53.7 | 87.2 | 43.3 | 89.6 | 66.9 | 39.2 | 88.2 | 43.4 | 34.97 | 31.7 | 47.71 | 61.6 | — | |
| TridentAdaptBackbone=ResNet-101, Architecture=DeepLab-v22021.11 | 53.3 | 91.3 | 51.5 | 86.4 | 38.8 | 36.4 | 42.3 | 45.4 | 42 | 86.6 | 36.4 | 84.3 | 67.7 | 42.8 | 89.1 | 41.7 | 38.2 | 20.6 | 40.3 | 30.7 | — | |
| DPL-DualBackbone=ResNet-1012021.08 | 53.3 | 92.8 | 54.4 | 86.2 | 41.6 | 32.7 | 36.4 | 49 | 34 | 85.8 | 41.3 | 86 | 63.2 | 34.2 | 87.2 | 39.3 | 44.5 | 18.7 | 42.6 | 43.1 | — | |
| FCNadv + RPT3 + MSDomain Adaptation Strategy=representation-level domain adaptation by adversarial learning, Multi-scale (MS)=true2020.06 | 53.2 | 89.7 | 44.8 | 86.4 | 44.2 | 30.6 | 41.4 | 51.7 | 33 | 87.8 | 39.4 | 86.3 | 65.6 | 24.5 | 89 | 36.2 | 46.8 | 17.6 | 39.1 | 58.3 | — | |
| DPLBackbone=ResNet-1012021.08 | 52.8 | 92.5 | 52.8 | 86 | 38.5 | 31.7 | 36.2 | 47.3 | 34.9 | 85.5 | 39.9 | 85.2 | 62.9 | 33.9 | 86.8 | 37.2 | 45.3 | 20.1 | 44.1 | 42.4 | — | |
| FCNadv + RPT3Domain Adaptation Strategy=representation-level domain adaptation by adversarial learning2020.06 | 52.6 | 89.2 | 43.3 | 86.1 | 39.5 | 29.9 | 40.2 | 49.6 | 33.1 | 87.4 | 38.5 | 86 | 64.4 | 25.1 | 88.5 | 36.6 | 45.8 | 23.9 | 36.5 | 56.8 | — | |
| DACSSource Domain=GTA2026.05 | 52.1 | 89.9 | 39.7 | 87.9 | 30.7 | 39.5 | 38.5 | 46.4 | 52.8 | 88 | 44 | 88.8 | 67.2 | 35.8 | 84.5 | 45.7 | 50.2 | 0 | 27.3 | 34 | — | |
| DT-STAdaptation Setting=SFDA, Backpropagation (BP)=true, Backbone=ResNet-101, Architecture=DeepLabV22026.07 | 52.1 | 90.3 | 47.8 | 84.3 | 38.8 | 22.7 | 32.4 | 41.8 | 41.2 | 85.8 | 42.5 | 87.8 | 62.6 | 37 | 82.5 | 25.8 | 32 | 29.8 | 48 | 56.9 | — | |
| TPLDBackbone=ResNet-1012021.08 | 51.2 | 94.2 | 60.5 | 82.8 | 36.6 | 16.6 | 39.3 | 29 | 25.5 | 85.6 | 44.9 | 84.4 | 60.6 | 27.4 | 84.1 | 37 | 47 | 31.2 | 36.1 | 50.3 | — | |
| FDA-MBTBackbone=ResNet-1012021.08 | 50.5 | 92.5 | 53.3 | 82.4 | 26.5 | 27.6 | 36.4 | 40.6 | 38.9 | 82.3 | 39.8 | 78 | 62.6 | 34.4 | 84.9 | 34.1 | 53.1 | 16.9 | 27.7 | 46.4 | — | |
| FCNadv + RPT1Domain Adaptation Strategy=representation-level domain adaptation by adversarial learning2020.06 | 50.4 | 88.7 | 37 | 85.2 | 36.6 | 27.7 | 42.6 | 49.1 | 30 | 86.9 | 37.6 | 80.7 | 66.8 | 27.5 | 88.1 | 30.3 | 39.5 | 22.5 | 28 | 53 | — | |
| SAITBackbone=ResNet-101, Architecture=DeepLab-v22021.11 | 50.4 | 91.2 | 43.3 | 85.2 | 38.6 | 25.9 | 34.7 | 41.3 | 41 | 85.5 | 46 | 86.5 | 61.7 | 33.8 | 85.5 | 34.4 | 48.7 | 0 | 36.1 | 37.8 | — | |
| CAG-UDABackbone=ResNet-101, Adaptation Protocol=UDA2019.10 | 50.2 | 90.4 | 51.6 | 83.8 | 34.2 | 27.8 | 38.4 | 25.3 | 48.4 | 85.4 | 38.2 | 78.1 | 58.6 | 34.6 | 84.7 | 21.9 | 42.7 | 41.1 | 29.3 | 37.2 | — | |
| TIRBackbone=ResNet-101, Architecture=DeepLab-v22021.11 | 50.2 | 92.9 | 55 | 85.3 | 34.2 | 31.1 | 34.9 | 40.7 | 34 | 85.2 | 40.1 | 87.1 | 61 | 31.1 | 82.5 | 32.3 | 42.9 | 0.3 | 36.4 | 46.1 | — | |
| Kim et al.Backbone=ResNet-1012021.08 | 50.2 | 92.9 | 55 | 85.3 | 34.2 | 31.1 | 34.9 | 40.7 | 34 | 85.2 | 40.1 | 87.1 | 61 | 31.1 | 82.5 | 32.3 | 42.9 | 0.3 | 36.4 | 46.1 | — | |
| Label-DrivenBackbone=ResNet-1012021.08 | 49.5 | 90.8 | 41.4 | 84.7 | 35.1 | 27.5 | 31.2 | 38 | 32.8 | 85.6 | 42.1 | 84.9 | 59.6 | 34.4 | 85 | 42.8 | 52.7 | 3.4 | 30.9 | 38.1 | — | |
| SDFAdaptation Setting=SFDA, Backpropagation (BP)=true, Backbone=ResNet-101, Architecture=DeepLabV22026.07 | 49.4 | 95.2 | 40.6 | 85.2 | 30.6 | 26.1 | 35.8 | 34.7 | 32.8 | 85.3 | 41.7 | 79.5 | 61 | 28.2 | 86.5 | 41.2 | 45.3 | 15.6 | 33.1 | 40 | — | |
| SIMBackbone=ResNet-1012020.03 | 49.2 | 90.6 | 44.7 | 84.8 | 34.3 | 28.7 | 31.6 | 35 | 37.6 | 84.7 | 43.3 | 85.3 | 57 | 31.5 | 83.8 | 42.6 | 48.5 | 1.9 | 30.4 | 39 | — | |
| SIMBackbone=ResNet-1012021.08 | 49.2 | 90.6 | 44.7 | 84.8 | 34.3 | 28.7 | 31.6 | 35 | 37.6 | 84.7 | 43.3 | 85.3 | 57 | 31.5 | 83.8 | 42.6 | 48.5 | 1.9 | 30.4 | 39 | — | |
| FADABackbone=ResNet-1012021.08 | 49.2 | 92.5 | 47.5 | 85.1 | 37.6 | 32.8 | 33.4 | 33.8 | 18.4 | 85.3 | 37.7 | 83.5 | 63.2 | 39.7 | 87.5 | 32.9 | 47.8 | 1.6 | 34.9 | 39.5 | — | |
| SISC+PWLSource mIoU=35.4, Adaptation Type=UDA2019.09 | 49 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 13.6 | |
| SISC-PWLAppr.=ST2019.09 | 49 | 89 | 45.2 | 78.2 | 22.9 | 27.3 | 37.4 | 46.1 | 43.8 | 82.9 | 18.6 | 61.2 | 60.4 | 26.7 | 85.4 | 35.9 | 44.9 | 36.4 | 37.2 | 49.3 | — | |
| MLSLDomain Adaptation Strategy=self-learning2020.06 | 49 | 89 | 45.2 | 78.2 | 22.9 | 27.3 | 37.4 | 46.1 | 43.8 | 82.9 | 18.6 | 61.2 | 60.4 | 26.7 | 85.4 | 35.9 | 44.9 | 36.4 | 37.2 | 49.3 | — | |
| SISCSource mIoU=35.4, Adaptation Type=UDA2019.09 | 48.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 13.3 | |
| SISCAppr.=ST2019.09 | 48.7 | 91 | 49.3 | 79.9 | 24.4 | 27.9 | 37.9 | 45.1 | 45.1 | 81.3 | 19 | 61.7 | 63.9 | 28 | 86.5 | 23.9 | 42.3 | 41.9 | 33.1 | 44.4 | — | |
| BDLBackbone=ResNet101, Oracle mIoU=65.12019.04 | 48.5 | 91 | 44.7 | 84.2 | 34.6 | 27.6 | 30.2 | 36 | 36 | 85 | 43.6 | 83 | 58.6 | 31.6 | 83.3 | 35.3 | 49.7 | 3.3 | 28.8 | 35.6 | — | |
| BLFBackbone=ResNet-101, Adaptation Protocol=UDA2019.10 | 48.5 | 91 | 44.7 | 84.2 | 34.6 | 27.6 | 30.2 | 36 | 36 | 85 | 43.6 | 83 | 58.6 | 31.6 | 83.3 | 35.3 | 49.7 | 3.3 | 28.8 | 35.6 | — | |
| Li et al.Backbone=ResNet-1012020.03 | 48.5 | 91 | 44.7 | 84.2 | 34.6 | 27.6 | 30.2 | 36 | 36 | 85 | 43.6 | 83 | 58.6 | 31.6 | 83.3 | 35.3 | 49.7 | 3.3 | 28.8 | 35.6 | — | |
| BidirectionalDomain Adaptation Strategy=others2020.06 | 48.5 | 91 | 44.7 | 84.2 | 34.6 | 27.6 | 30.2 | 36 | 36 | 85 | 43.6 | 83 | 58.6 | 31.6 | 83.3 | 35.3 | 49.7 | 3.3 | 28.8 | 35.6 | — | |
| BDLBackbone=ResNet-101, Architecture=DeepLab-v22021.11 | 48.5 | 91 | 44.7 | 84.2 | 34.6 | 27.6 | 30.2 | 36 | 36 | 85 | 43.6 | 83 | 58.6 | 31.6 | 83.3 | 35.3 | 49.7 | 3.3 | 28.8 | 35.6 | — | |
| BDLBackbone=ResNet-1012021.08 | 48.5 | 91 | 44.7 | 84.2 | 34.6 | 27.6 | 30.2 | 36 | 36 | 85 | 43.6 | 83 | 58.6 | 31.6 | 83.3 | 35.3 | 49.7 | 3.3 | 28.8 | 35.6 | — | |
| CBST + ParetoDABackbone=ResNet-1012021.12 | 48.1 | 82.4 | 50.7 | 80.2 | 34.9 | 36.7 | 45.9 | 47.2 | 45.1 | 72.9 | 27.8 | 46 | 62.3 | 33 | 81.4 | 29.5 | 44.3 | 14 | 29.8 | 49.3 | — | |
| HCLAdaptation Setting=SFDA, Backpropagation (BP)=true, Backbone=ResNet-101, Architecture=DeepLabV22026.07 | 48.1 | 92 | 55 | 80.4 | 33.5 | 24.6 | 37.1 | 35.1 | 28.8 | 83 | 37.6 | 82.3 | 59.4 | 27.6 | 83.6 | 32.3 | 36.6 | 14.1 | 28.7 | 43 | — | |
| PyCDANetwork=ResNet-382019.08 | 48 | 92.3 | 49.2 | 84.4 | 33.4 | 30.2 | 33.3 | 37.1 | 35.2 | 86.5 | 36.9 | 77.3 | 63.3 | 30.5 | 86.6 | 34.5 | 40.7 | 7.9 | 17.6 | 35.5 | — | |
| PyCDADomain Adaptation Strategy=self-learning2020.06 | 48 | 92.3 | 49.2 | 84.4 | 33.4 | 30.2 | 33.3 | 37.1 | 35.2 | 86.5 | 36.9 | 77.3 | 63.3 | 30.5 | 86.6 | 34.5 | 40.7 | 7.9 | 17.6 | 35.5 | — | |
| StylizationDomain Adaptation Strategy=appearance-level domain adaptation by image translation2020.06 | 47.8 | 86.9 | 44.5 | 84.7 | 38.8 | 26.6 | 32.1 | 42.3 | 22.5 | 84.7 | 30.9 | 85.9 | 67 | 28.1 | 85.7 | 38.3 | 31.8 | 21.5 | 31.3 | 24.6 | — | |
| PyCDANetwork=ResNet-1012019.08 | 47.4 | 90.5 | 36.3 | 84.4 | 32.4 | 28.7 | 34.6 | 36.4 | 31.5 | 86.8 | 37.9 | 78.5 | 62.3 | 21.5 | 85.6 | 27.9 | 34.8 | 18 | 22.9 | 49.3 | — | |
| Lian et al.Backbone=ResNet-1012020.03 | 47.4 | 90.5 | 36.3 | 84.4 | 32.4 | 28.7 | 34.6 | 36.4 | 31.5 | 86.8 | 37.9 | 78.5 | 62.3 | 21.5 | 85.6 | 27.9 | 34.8 | 18 | 22.9 | 49.3 | — | |
| PyCDABackbone=ResNet-101, Architecture=DeepLab-v22021.11 | 47.4 | 90.5 | 36.3 | 84.4 | 32.4 | 28.7 | 34.6 | 36.4 | 31.5 | 86.8 | 37.9 | 78.5 | 62.3 | 21.5 | 85.6 | 27.9 | 34.8 | 18 | 22.9 | 49.3 | — | |
| FDA + ParetoDABackbone=ResNet-1012021.12 | 47.2 | 85.4 | 37.6 | 84 | 32 | 23.8 | 33.1 | 39.6 | 25.9 | 84 | 30.6 | 81.6 | 61.3 | 30.9 | 85.4 | 31.9 | 45 | 7.6 | 32.4 | 45.4 | — | |
| CRSTBackbone=ResNet-101, Architecture=DeepLab-v22021.11 | 47.1 | 91 | 55.4 | 80 | 33.7 | 21.4 | 37.3 | 32.9 | 24.5 | 85 | 34.1 | 80.8 | 57.7 | 24.6 | 84.1 | 27.8 | 30.1 | 26.9 | 26 | 42.3 | — | |
| CBSTBackbone Network=ResNet-38, Adaptation=true, Mechanism=Self-Training2018.12 | 47 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 11.6 | |
| Zou et al.Backbone=ResNet-1012020.03 | 47 | 89.6 | 58.9 | 78.5 | 33 | 22.3 | 41.4 | 48.2 | 39.2 | 83.6 | 24.3 | 65.4 | 49.3 | 20.2 | 83.3 | 39 | 48.6 | 12.5 | 20.3 | 35.3 | — | |
| CBSTDomain Adaptation Strategy=self-learning2020.06 | 47 | 89.6 | 58.9 | 78.5 | 33 | 22.3 | 41.4 | 48.2 | 39.2 | 83.6 | 24.3 | 65.4 | 49.3 | 20.2 | 83.3 | 39 | 48.6 | 12.5 | 20.3 | 35.3 | — | |
| FCANBackbone Network=ResNet-101, Adaptation=true, Mechanism=Adversary2018.12 | 46.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 17.4 | |
| FCANNetwork=ResNet-101, image_size=original2019.08 | 46.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FCANDomain Adaptation Strategy=appearance-level + representation-level adaptation2020.06 | 46.6 | 88.9 | 37.9 | 82.9 | 33.2 | 26.1 | 42.8 | 43.2 | 28.4 | 86.5 | 35.2 | 78 | 65.9 | 22.8 | 86.7 | 23.7 | 34.9 | 2.7 | 24 | 41.9 | — | |
| patch-level alignmentBackbone=ResNet-1012019.01 | 46.5 | 92.3 | 51.9 | 82.1 | 29.2 | 25.1 | 24.5 | 33.8 | 33 | 82.4 | 32.8 | 82.2 | 58.6 | 27.2 | 84.3 | 33.4 | 46.3 | 2.2 | 29.5 | 32.3 | — | |
| DPL-DualBackbone=VGG-162021.08 | 46.5 | 89.2 | 44 | 83.5 | 35 | 24.7 | 27.8 | 38.3 | 25.3 | 84.2 | 39.5 | 81.6 | 54.7 | 25.8 | 83.3 | 29.3 | 49 | 5.2 | 30.2 | 32.6 | — | |
| Chen et al.Backbone=ResNet-1012020.03 | 46.4 | 89.4 | 43 | 82.1 | 30.5 | 21.3 | 30.3 | 34.7 | 24 | 85.3 | 39.4 | 78.2 | 63 | 22.9 | 84.6 | 36.4 | 43 | 5.5 | 34.7 | 33.5 | — | |
| IntraDA (Ours)Backbone=ResNet-101, Source Domain=GTA5, Adaptation Framework=AdvEnt [29]2020.04 | 46.3 | 90.6 | 36.1 | 82.6 | 29.5 | 21.3 | 27.6 | 31.4 | 23.1 | 85.2 | 39.3 | 80.2 | 59.3 | 29.4 | 86.4 | 33.6 | 53.9 | 0 | 32.7 | 37.6 | — | |
| IntroDABackbone=ResNet-101, Architecture=DeepLab-v22021.11 | 46.3 | 90.6 | 36.1 | 82.6 | 29.5 | 21.3 | 27.6 | 31.4 | 23.1 | 85.2 | 39.3 | 80.2 | 59.3 | 29.4 | 86.4 | 33.6 | 53.9 | 0 | 32.7 | 37.6 | — | |
| CBSTSource mIoU=35.4, Adaptation Type=UDA2019.09 | 46.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 10.8 | |
| CBST-SPAppr.=ST2019.09 | 46.2 | 88 | 56.2 | 77 | 27.4 | 22.4 | 40.7 | 47.3 | 40.9 | 82.4 | 21.6 | 60.3 | 50.2 | 20.4 | 83.8 | 35 | 51 | 15.2 | 20.6 | 37 | — | |
| DPLBackbone=VGG-162021.08 | 46.2 | 88.9 | 43.6 | 83.4 | 33.8 | 24.7 | 28 | 37.6 | 26.2 | 84.1 | 40.3 | 81.5 | 54.9 | 25 | 83 | 27.7 | 48.6 | 4.8 | 29.1 | 32 | — | |
| AdvEnt + ParetoDABackbone=ResNet-1012021.12 | 46.1 | 79.6 | 29.1 | 83.4 | 30.4 | 22.4 | 32.6 | 39.2 | 26.7 | 82.5 | 26.5 | 82.2 | 63 | 29.8 | 85.9 | 36.1 | 43.5 | 9.2 | 29.9 | 43.7 | — | |
| SFDAAdaptation Setting=SFDA, Backpropagation (BP)=true, Backbone=ResNet-101, Architecture=DeepLabV22026.07 | 45.8 | 91.7 | 52.7 | 82.2 | 28.7 | 20.3 | 36.5 | 30.6 | 23.6 | 81.7 | 35.6 | 84.8 | 59.5 | 22.6 | 83.4 | 29.6 | 32.4 | 11.8 | 23.8 | 39.6 | — | |
| AdvEnt+MinEntBackbone=ResNet-101, Approach=Adv + Ent2018.11 | 45.5 | 89.4 | 33.1 | 81 | 26.6 | 26.8 | 27.2 | 33.5 | 24.7 | 83.9 | 36.7 | 78.8 | 58.7 | 30.5 | 84.8 | 38.5 | — | — | 31.6 | 32.4 | — | |
| AdvEntBackbone=ResNet-101, Adaptation Protocol=UDA2019.10 | 45.5 | 89.4 | 33.1 | 81 | 26.6 | 26.8 | 27.2 | 33.5 | 24.7 | 83.9 | 36.7 | 78.8 | 58.7 | 30.5 | 84.8 | 38.5 | 44.5 | 1.7 | 31.6 | 32.4 | — | |
| Vu et al.Backbone=ResNet-1012020.03 | 45.5 | 89.4 | 33.1 | 81 | 26.6 | 26.8 | 27.2 | 33.5 | 24.7 | 83.9 | 36.7 | 78.8 | 58.7 | 30.5 | 84.8 | 38.5 | 44.5 | 1.7 | 31.6 | 32.4 | — | |
| ADVENTDomain Adaptation Strategy=representation-level domain adaptation by adversarial learning2020.06 | 45.5 | 89.4 | 33.1 | 81 | 26.6 | 26.8 | 27.2 | 33.5 | 24.7 | 83.9 | 36.7 | 78.8 | 58.7 | 30.5 | 84.8 | 38.5 | 44.5 | 1.7 | 31.6 | 32.4 | — | |
| ADVENTSource Domain=GTA2026.05 | 45.5 | 89.4 | 33.1 | 81 | 26.6 | 26.8 | 27.2 | 33.5 | 24.7 | 83.9 | 36.7 | 78.8 | 58.7 | 30.5 | 84.8 | 38.5 | 44.5 | 1.7 | 31.6 | 32.4 | — | |
| DISEBase Model=Deeplab v22019.03 | 45.4 | 91.5 | 47.5 | 82.5 | 31.3 | 25.6 | 33 | 33.7 | 25.8 | 82.7 | 28.8 | 82.7 | 62.4 | 30.8 | 85.2 | 27.7 | 34.5 | 640 | 25.2 | 24.4 | — | |
| All StructureSource mIoU=36.6, Adaptation Type=UDA2019.09 | 45.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 8.8 | |
| All StructureAppr.=Adv2019.09 | 45.4 | 91.5 | 47.5 | 82.5 | 31.3 | 25.6 | 33 | 33.7 | 25.8 | 82.7 | 28.8 | 82.7 | 62.4 | 30.8 | 85.2 | 27.7 | 34.5 | 6.4 | 25.2 | 24.4 | — | |
| DISEBackbone=ResNet-101, Adaptation Protocol=UDA2019.10 | 45.4 | 91.5 | 47.5 | 82.5 | 31.3 | 25.6 | 33 | 33.7 | 25.8 | 82.7 | 28.8 | 82.7 | 62.4 | 30.8 | 85.2 | 27.7 | 34.5 | 6.4 | 25.2 | 24.4 | — | |
| Chang et al.Backbone=ResNet-1012020.03 | 45.4 | 91.5 | 47.5 | 82.5 | 31.3 | 25.6 | 33 | 33.7 | 25.8 | 82.7 | 28.8 | 82.7 | 62.4 | 30.8 | 85.2 | 27.7 | 34.5 | 6.4 | 25.2 | 24.4 | — | |
| Du et al.Backbone=ResNet-1012020.03 | 45.4 | 90.3 | 38.9 | 81.7 | 24.8 | 22.9 | 30.5 | 37 | 21.2 | 84.8 | 38.8 | 76.9 | 58.8 | 30.7 | 85.7 | 30.6 | 38.1 | 5.9 | 28.3 | 36.9 | — | |
| SSF-DANDomain Adaptation Strategy=representation-level domain adaptation by adversarial learning2020.06 | 45.4 | 90.3 | 38.9 | 81.7 | 24.8 | 22.9 | 30.5 | 37 | 21.2 | 84.8 | 38.8 | 76.9 | 58.8 | 30.7 | 85.7 | 30.6 | 38.1 | 5.9 | 28.3 | 36.9 | — | |
| DISEDomain Adaptation Strategy=appearance-level + representation-level adaptation2020.06 | 45.4 | 91.5 | 47.5 | 82.5 | 31.3 | 25.6 | 33 | 33.7 | 25.8 | 82.7 | 28.8 | 82.7 | 62.4 | 30.8 | 85.2 | 27.7 | 34.5 | 6.4 | 25.2 | 24.4 | — | |
| ADVENTBackbone=ResNet-101, Architecture=DeepLab-v22021.11 | 45.4 | 89.4 | 33.1 | 81 | 26.6 | 26.8 | 27.2 | 33.5 | 24.7 | 83.9 | 36.7 | 78.8 | 58.7 | 30.5 | 84.8 | 38.5 | 44.5 | 1.7 | 31.6 | 32.4 | — | |
| SSF-DANBackbone=ResNet-101, Architecture=DeepLab-v22021.11 | 45.4 | 90.3 | 38.9 | 81.7 | 24.8 | 22.9 | 30.5 | 37 | 21.2 | 84.8 | 38.8 | 76.9 | 58.8 | 30.7 | 85.7 | 30.6 | 38.1 | 5.9 | 28.3 | 36.9 | — | |
| CBSTNetwork=ResNet-382019.08 | 45.2 | 86.8 | 46.7 | 76.9 | 26.3 | 24.8 | 42 | 46 | 38.6 | 80.7 | 15.7 | 48 | 57.3 | 27.9 | 78.2 | 24.5 | 49.6 | 17.7 | 25.5 | 45.1 | — | |
| MaxSquareDomain Adaptation Strategy=others2020.06 | 45.2 | 89.3 | 40.5 | 81.2 | 29 | 20.4 | 25.6 | 34.4 | 19 | 83.6 | 34.4 | 76.5 | 59.2 | 27.4 | 83.8 | 38.4 | 43.6 | 7.1 | 32.2 | 32.5 | — | |
| CBSTBackbone=ResNet-1012021.12 | 45.2 | 86.8 | 46.7 | 76.9 | 26.3 | 24.8 | 42 | 46 | 38.6 | 80.7 | 15.7 | 48 | 57.3 | 27.9 | 78.2 | 24.5 | 49.6 | 17.7 | 25.5 | 45.1 | — | |
| URMDAAdaptation Setting=SFDA, Backpropagation (BP)=true, Backbone=ResNet-101, Architecture=DeepLabV22026.07 | 45.1 | 92.3 | 55.2 | 81.6 | 30.8 | 18.8 | 37.1 | 17.7 | 12.1 | 84.2 | 35.9 | 83.8 | 57.7 | 24.1 | 81.7 | 27.5 | 44.3 | 6.9 | 24.1 | 40.4 | — | |
| ADVENTNetwork=ResNet-1012019.08 | 44.8 | 87.6 | 21.4 | 82 | 34.8 | 26.2 | 28.5 | 35.6 | 23 | 84.5 | 35.1 | 76.2 | 58.6 | 30.7 | 84.8 | 34.2 | 43.4 | 0.4 | 28.4 | 35.3 | — | |
| FDABackbone=ResNet-1012021.12 | 44.6 | 90 | 40.5 | 79.4 | 25.3 | 26.7 | 30.6 | 31.9 | 29.3 | 79.4 | 28.8 | 76.5 | 56.4 | 27.5 | 81.7 | 27.7 | 45.1 | 17 | 23.8 | 29.6 | — | |
| Swin-B ViTBackbone=Swin-B ViT, Parameters=88M, Source Domain=GTAV2022.04 | 44.6 | 63.3 | 28.6 | 68.3 | 16.8 | 23.4 | 37.8 | 51 | 34.3 | 83.8 | 42.1 | 85.7 | 68.5 | 25.4 | 83.5 | 36.3 | 17.7 | 2.9 | 36.1 | 42.3 | — | |
| CGANBackbone Network=VGG-16, Adaptation=true, Mechanism=Adversary2018.12 | 44.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 23.4 | |
| SWDBackbone=ResNet-1012019.03 | 44.5 | 92 | 46.4 | 82.4 | 24.8 | 24 | 35.1 | 33.4 | 34.2 | 83.6 | 30.4 | 80.9 | 56.9 | 21.9 | 82 | 24.4 | 28.7 | 6.1 | 25 | 33.6 | — | |
| Hong et al.Base Model=FCN8s2019.03 | 44.5 | 89.2 | 49 | 70.7 | 13.5 | 10.9 | 38.5 | 29.4 | 33.7 | 77.9 | 37.6 | 65.8 | 75.1 | 32.4 | 77.8 | 39.2 | 45.2 | 0 | 25.5 | 35.4 | — | |
| Hong et al.Backbone=ResNet-1012020.03 | 44.5 | 89.2 | 49 | 70.7 | 13.5 | 10.9 | 38.5 | 29.4 | 33.7 | 77.9 | 37.6 | 65.8 | 75.1 | 32.4 | 77.8 | 39.2 | 45.2 | 0 | 25.5 | 35.4 | — | |
| ConditionalDomain Adaptation Strategy=representation-level domain adaptation by adversarial learning2020.06 | 44.5 | 89.2 | 49 | 70.7 | 13.5 | 10.9 | 38.5 | 29.4 | 33.7 | 77.9 | 37.6 | 65.8 | 75.1 | 32.4 | 77.8 | 39.2 | 45.2 | 0 | 25.5 | 35.4 | — | |
| CGANNetwork=VGG-19, Train w/ Tgt=true, Val on Tgt=false2019.09 | 44.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SegF. MiT-B5Backbone=MiT-B5, Parameters=84.7M, Source Domain=GTAV2022.04 | 44.5 | 77.1 | 15.2 | 83.8 | 30.8 | 32 | 27.9 | 41.5 | 18.5 | 86.5 | 42.5 | 86.8 | 62.6 | 22.2 | 87 | 42.7 | 36.8 | 6.1 | 33.5 | 12.5 | — | |
| AdvEntBackbone=ResNet-101, Approach=Adversarial Training2018.11 | 43.8 | 89.9 | 36.5 | 81.6 | 29.2 | 25.2 | 28.5 | 32.3 | 22.4 | 83.9 | 34 | 77.1 | 57.4 | 27.9 | 83.7 | 29.4 | — | — | 28.4 | 23.3 | — | |
| AdvEntBackbone=ResNet-101, Source Domain=GTA5, Adaptation Framework=AdvEnt [29]2020.04 | 43.8 | 89.9 | 36.5 | 81.6 | 29.2 | 25.2 | 28.5 | 32.3 | 22.4 | 83.9 | 34 | 77.1 | 57.4 | 27.9 | 83.7 | 29.4 | 39.1 | 1.5 | 28.4 | 23.3 | — | |
| AdvEntBackbone=ResNet-1012021.12 | 43.8 | 89.9 | 36.5 | 81.6 | 29.2 | 25.2 | 28.5 | 32.3 | 22.4 | 83.9 | 34 | 77.1 | 57.4 | 27.9 | 83.7 | 29.4 | 39.1 | 1.5 | 28.4 | 23.3 | — |