Semantic Segmentation on Synthia to Cityscapes (test)
93.3Road IoUCorDA
Evaluation results
| Method | Links | ||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CorDACategory=UDA with auxiliary tasks, Uses source depth labels=true2022.11 | 93.3 | 55 | 61.6 | 85.3 | 19.6 | 5.1 | 37.8 | 36.6 | 42.8 | 84.9 | 90.4 | 69.7 | 41.8 | 85.6 | 38.4 | 32.6 | 53.9 | — | — | — | — | — | |
| CorDAUDA type=Auxiliary task based2022.11 | 93.3 | 55 | 61.6 | 85.3 | 19.6 | 5.1 | 37.8 | 36.6 | 42.8 | 84.9 | 90.4 | 69.7 | 41.8 | 85.6 | 38.4 | 32.6 | 53.9 | — | — | — | — | — | |
| BiMaL2021.10 | 92.8 | — | 51.5 | 81.5 | 10.2 | 1 | 30.4 | 17.6 | 15.9 | 82.4 | 84.6 | 55.9 | 22.3 | 85.7 | 44.5 | 24.6 | 38.8 | 46.2 | 53.7 | — | — | — | |
| MetaCorrectionBackbone=ResNet-101, Parameters=43M2022.03 | 92.6 | 45.1 | 52.7 | 81.3 | 8.9 | 2.4 | 28.1 | 13 | 7.3 | 83.5 | 85 | 60.1 | 19.7 | 84.8 | 37.2 | 21.5 | 43.9 | — | — | 52.5 | — | — | |
| LTIRmulti-stage training=false, multi-model fusion=false2022.04 | 92.6 | — | 53.2 | 79.2 | — | — | — | 1.6 | 7.5 | 78.6 | 84.4 | 52.6 | 20 | 82.1 | 34.8 | 14.6 | 39.4 | — | — | 49.3 | — | — | |
| MetaCorrectmulti-stage training=false, multi-model fusion=false2022.04 | 92.6 | 45.1 | 52.7 | 81.3 | 8.9 | 240 | 28.1 | 13 | 7.3 | 83.5 | 85 | 60.1 | 19.7 | 84.8 | 37.2 | 21.5 | 43.9 | — | — | 52.5 | — | — | |
| ELDACategory=UDA with auxiliary tasks, Uses source depth labels=false2022.11 | 92.6 | 55.2 | 56.6 | 85.5 | 24.2 | 2.1 | 37.6 | 38.1 | 43.1 | 85.7 | 91.5 | 69.8 | 42 | 87.2 | 47.6 | 20 | 50.1 | — | — | — | — | — | |
| ELDAUDA type=Auxiliary task based (edge information)2022.11 | 92.6 | 55.2 | 56.6 | 85.5 | 24.2 | 2.1 | 37.6 | 38.1 | 43.1 | 85.7 | 91.5 | 69.8 | 42 | 87.2 | 47.6 | 20 | 50.1 | — | — | — | — | — | |
| RCCR2021.10 | 92.5 | — | 58.7 | 83.7 | 15.1 | 1.3 | 34.7 | 26.6 | 27.1 | 82.6 | 87.3 | 66 | 34.9 | 86.5 | 50.5 | 23.6 | 47.4 | 51.1 | 59 | — | — | — | |
| WLabel2021.10 | 92 | — | 53.5 | 80.9 | 11.4 | 0.4 | 21.8 | 3.8 | 6 | 81.6 | 84.4 | 60.8 | 24.4 | 80.5 | 39 | 26 | 41.7 | 44.3 | 51.9 | — | — | — | |
| DISEBackbone=ResNet-101, Architecture=DeepLab-V2, Training Strategy=Adversarial Training2020.10 | 91.7 | — | 53.5 | 77.1 | 2.5 | 0.2 | 27.1 | 6.2 | 7.6 | 78.4 | 81.2 | 55.8 | 19.2 | 82.3 | 30.3 | 17.1 | 34.3 | 41.5 | 48.7 | — | — | — | |
| ASA2021.10 | 91.2 | — | 48.5 | 80.4 | 3.7 | 0.3 | 21.7 | 5.5 | 5.2 | 79.5 | 83.6 | 56.4 | 21.9 | 80.3 | 36.2 | 20 | 32.9 | 41.7 | 49.3 | — | — | — | |
| ASABackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 91.2 | 41.7 | 48.5 | 80.4 | 3.7 | 0.3 | 21.7 | 5.5 | 5.2 | 79.5 | 83.6 | 56.4 | 21 | 80.3 | 36.2 | 20 | 32.9 | — | — | 49.3 | — | — | |
| TransDA-BBackbone=Swin-B ViT, Parameters=88M2022.03 | 90.4 | 59.3 | 54.8 | 86.4 | 31.1 | 1.7 | 53.8 | 61.1 | 37.1 | 90.3 | 93 | 71.2 | 25.3 | 92.3 | 66 | 44.4 | 49.8 | — | — | 66.3 | — | — | |
| TGCF-DAArchitecture=VGG162020.03 | 90.1 | 38.5 | 48.6 | 80.7 | 2.2 | 0.2 | 27.2 | 3.2 | 14.3 | 82.1 | 78.4 | 54.4 | 16.4 | 82.5 | 12.3 | 1.7 | 21.8 | — | — | — | — | — | |
| SACBackbone=ResNet-101, Parameters=43M2022.03 | 89.3 | 52.6 | 47.2 | 85.5 | 26.5 | 1.3 | 43 | 45.5 | 32 | 87.1 | 89.3 | 63.6 | 25.4 | 86.9 | 35.6 | 30.4 | 53 | — | — | 59.3 | — | — | |
| DADAArchitecture=ResNet1012020.03 | 89.2 | 49.8 | 44.8 | 81.4 | — | — | — | 8.6 | 11.1 | 81.8 | 84 | 54.7 | 19.3 | 79.7 | 40.7 | 14 | 38.8 | — | — | — | — | — | |
| DADACategory=UDA with auxiliary tasks, Uses source depth labels=true2022.11 | 89.2 | 42.6 | 44.8 | 81.4 | 6.8 | 0.3 | 26.2 | 8.6 | 11.1 | 81.8 | 84 | 54.7 | 19.3 | 79.7 | 40.7 | 14 | 38.8 | — | — | — | — | — | |
| DADAUDA type=Auxiliary task based2022.11 | 89.2 | 42.6 | 44.8 | 81.4 | 6.8 | 0.3 | 26.2 | 8.6 | 11.1 | 81.8 | 84 | 54.7 | 19.3 | 79.7 | 40.7 | 14 | 38.8 | — | — | — | — | — | |
| DADABackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 89.2 | 42.6 | 44.8 | 81.4 | 6.8 | 0.3 | 26.2 | 8.6 | 11.1 | 81.8 | 84 | 54.7 | 19.3 | 79.7 | 40.7 | 14 | 38.8 | — | — | 49.8 | — | — | |
| TransDA-BBackbone=Swin-B (88M)2022.03 | 88.9 | 56.8 | 46.1 | 85.4 | 11.6 | 1.7 | 50.7 | 61.3 | 38.8 | 88.4 | 93.7 | 74.4 | 30 | 89.6 | 49.4 | 49.3 | 49.2 | — | — | 65 | — | — | |
| Chao et al.multi-stage training=false, multi-model fusion=true2022.04 | 88.7 | 52.6 | 46.7 | 83.8 | 22.7 | 410 | 35 | 35.9 | 36.1 | 82.8 | 81.4 | 61.6 | 32.1 | 87.9 | 52.8 | 32 | 57.7 | — | — | 60 | — | — | |
| MCDBackbone=DRN-105, k=42017.12 | 88.1 | 37.2 | 43.2 | 79.1 | 2.4 | 0.1 | 27.3 | 7.4 | 4.9 | 83.4 | 81.1 | 51.3 | 10.9 | 82.1 | 29 | 5.7 | 0 | — | — | — | — | — | |
| Maximum Classifier DiscrepancyBackbone=DRN-105, k=42017.12 | 88.1 | 37.2 | 43.2 | 79.1 | 2.4 | 0.1 | 27.3 | 7.4 | 4.9 | 83.4 | 81.1 | 51.3 | 10.9 | 82.1 | 29 | 5.7 | 0 | — | — | — | — | — | |
| GUDACategory=UDA with auxiliary tasks, Uses source depth labels=true2022.11 | 88.1 | 51 | 53 | 84 | 22 | 1.4 | 39.6 | 28.2 | 24.8 | 82.7 | 81.5 | 65.5 | 22.7 | 89.3 | 50.5 | 25.1 | 57.5 | — | — | — | — | — | |
| GUDAUDA type=Auxiliary task based2022.11 | 88.1 | 51 | 53 | 84 | 22 | 1.4 | 39.6 | 24.8 | 82.7 | 81.5 | 65.5 | 22.7 | 89.3 | 50.5 | 25.1 | 57.5 | 51 | — | — | — | — | — | |
| ProDABackbone=ResNet-101, Parameters=43M2022.03 | 87.8 | 55.5 | 45.7 | 84.6 | 37.1 | 0.6 | 44 | 54.6 | 37 | 88.1 | 84.4 | 74.2 | 24.3 | 88.2 | 51.1 | 40.5 | 45.6 | — | — | 62 | — | — | |
| ProDAmulti-stage training=true, multi-model fusion=false2022.04 | 87.8 | 55.5 | 45.7 | 84.6 | 37.1 | 60 | 44 | 54.6 | 37 | 88.1 | 84.4 | 74.2 | 24.3 | 88.2 | 51.1 | 40.5 | 45.6 | — | — | 62 | — | — | |
| Seg-UncertaintyBackbone=ResNet-101, Parameters=43M2022.03 | 87.6 | 47.9 | 41.9 | 83.1 | 14.7 | 1.7 | 36.2 | 31.3 | 19.9 | 81.6 | 80.6 | 63 | 21.8 | 86.2 | 40.7 | 23.6 | 53.1 | — | — | 54.9 | — | — | |
| UncertaintyCategory=Pure semantic segmentation based UDA2022.11 | 87.6 | 47.9 | 41.9 | 83.1 | 14.7 | 1.7 | 36.2 | 31.3 | 19.9 | 81.6 | 80.6 | 63 | 21.8 | 86.2 | 40.7 | 23.6 | 53.1 | — | — | — | — | — | |
| UncertaintyUDA type=Pure semantic segmentation2022.11 | 87.6 | 47.9 | 41.9 | 83.1 | 14.7 | 1.7 | 36.2 | 31.3 | 19.9 | 81.6 | 80.6 | 63 | 21.8 | 86.2 | 40.7 | 23.6 | 53.1 | — | — | — | — | — | |
| UncertaintyBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 87.6 | 47.9 | 41.9 | 83.1 | 14.7 | 1.7 | 36.2 | 31.3 | 19.9 | 81.6 | 80.6 | 63 | 21.8 | 86.2 | 40.7 | 23.6 | 53.1 | — | — | 54.9 | — | — | |
| DPLmulti-stage training=false, multi-model fusion=false2022.04 | 87.5 | 47 | 45.7 | 82.8 | 13.3 | 60 | 33.2 | 22 | 20.1 | 83.1 | 86 | 56.6 | 21.9 | 83.1 | 40.3 | 29.8 | 45.7 | — | — | 54.2 | — | — | |
| SAC2022.03 | 87.4 | 52.7 | 41 | 85.5 | 17.5 | 2.6 | 40.5 | 44.7 | 34.4 | 87.9 | 91.2 | 68 | 31 | 89.3 | 33.2 | 38.6 | 49.9 | — | — | 60.2 | — | — | |
| DAST2021.10 | 87.1 | — | 44.5 | 82.3 | 10.7 | 0.8 | 29.9 | 13.9 | 13.1 | 81.6 | 86 | 60.3 | 25.1 | 83.1 | 40.1 | 24.4 | 40.5 | 45.2 | 52.5 | — | — | — | |
| CTRLCategory=UDA with auxiliary tasks, Uses source depth labels=true2022.11 | 86.4 | 45 | 42.5 | 80.4 | 20 | 1 | 27.7 | 10.5 | 13.3 | 80.6 | 82.6 | 61 | 23.7 | 81.8 | 42.9 | 21 | 44.7 | — | — | — | — | — | |
| CTRLUDA type=Auxiliary task based2022.11 | 86.4 | 45 | 42.5 | 80.4 | 20 | 1 | 27.7 | 10.5 | 13.3 | 80.6 | 82.6 | 61 | 23.7 | 81.8 | 42.9 | 21 | 44.7 | — | — | — | — | — | |
| APODABackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 86.4 | — | 41.3 | 79.3 | — | — | — | 22.6 | 17.3 | 80.3 | 81.6 | 56.9 | 21 | 84.1 | 49.1 | 24.6 | 45.7 | — | — | 53.1 | — | — | |
| BDLBackbone=ResNet-101, Oracle mIoU=71.72019.04 | 86 | 51.4 | 46.7 | 80.3 | — | — | — | 14.1 | 11.6 | 79.2 | 81.3 | 54.1 | 27.9 | 73.7 | 42.2 | 25.7 | 45.3 | — | — | — | — | — | |
| BDLArchitecture=ResNet1012020.03 | 86 | 51.4 | 46.7 | 80.3 | — | — | — | 14.1 | 11.6 | 79.2 | 81.3 | 54.1 | 27.9 | 73.7 | 42.2 | 25.7 | 45.3 | — | — | — | — | — | |
| BDL2021.10 | 86 | — | 46.7 | 80.3 | — | — | — | 14.1 | 11.6 | 79.2 | 81.3 | 54.1 | 27.9 | 73.7 | 42.2 | 25.7 | 45.3 | — | 51.4 | — | — | — | |
| BLBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 86 | — | 46.7 | 80.3 | — | — | — | 14.1 | 11.6 | 79.2 | 81.3 | 54.1 | 27.9 | 73.7 | 42.2 | 25.7 | 45.3 | — | — | 51.4 | — | — | |
| ADVENTBackbone=ResNet-101, Architecture=DeepLab-V2, Training Strategy=Adversarial Training2020.10 | 85.6 | — | 42.2 | 79.7 | 8.7 | 0.4 | 25.9 | 5.4 | 8.1 | 80.4 | 84.1 | 57.9 | 23.8 | 73.3 | 36.4 | 14.2 | 33 | 41.2 | 48 | — | — | — | |
| AdvEntBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 85.6 | 41.2 | 42.2 | 79.7 | 8.7 | 0.4 | 25.9 | 5.4 | 8.1 | 80.4 | 84.1 | 57.9 | 23.8 | 73.3 | 36.4 | 14.2 | 33 | — | — | 48 | — | — | |
| Uncertainty + AdaBoost StudentBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 85.6 | 50.4 | 43.9 | 83.9 | 19.2 | 1.7 | 38 | 37.9 | 19.6 | 85.5 | 88.4 | 64.1 | 25.7 | 86.6 | 43.9 | 31.2 | 51.3 | — | — | 57.5 | — | — | |
| LDR2021.10 | 85.1 | — | 44.5 | 81 | — | — | — | 16.4 | 15.2 | 80.1 | 84.8 | 59.4 | 31.9 | 73.2 | 41 | 32.6 | 44.7 | — | 53.1 | — | — | — | |
| MCDBackbone=DRN-105, k=32017.12 | 84.8 | 37.3 | 43.6 | 79 | 3.9 | 0.2 | 29.1 | 7.2 | 5.5 | 83.8 | 83.1 | 51 | 11.7 | 79.9 | 27.2 | 6.2 | 0 | — | — | — | — | — | |
| Maximum Classifier DiscrepancyBackbone=DRN-105, k=32017.12 | 84.8 | 37.3 | 43.6 | 79 | 3.9 | 0.2 | 29.1 | 7.2 | 5.5 | 83.8 | 83.1 | 51 | 11.7 | 79.9 | 27.2 | 6.2 | 0 | — | — | — | — | — | |
| SP-AdvBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 84.8 | — | 35.8 | 78.6 | — | — | — | 6.2 | 15.6 | 80.5 | 82 | 66.5 | 22.7 | 74.3 | 34.1 | 19.2 | 27.3 | — | — | 48.3 | — | — | |
| CAG-UDACategory=Pure semantic segmentation based UDA2022.11 | 84.7 | 44.5 | 40.8 | 81.7 | 7.8 | 0 | 35.1 | 13.3 | 22.7 | 84.5 | 77.6 | 64.2 | 27.8 | 80.9 | 19.7 | 22.7 | 48.3 | — | — | — | — | — | |
| CAG-UDAUDA type=Pure semantic segmentation2022.11 | 84.7 | 44.5 | 40.8 | 81.7 | 7.8 | 0 | 35.1 | 13.3 | 22.7 | 84.5 | 77.6 | 64.2 | 27.8 | 80.9 | 19.7 | 22.7 | 48.3 | — | — | — | — | — | |
| SSF-DANBackbone=ResNet-101, Architecture=DeepLab-V2, Training Strategy=Adversarial Training2020.10 | 84.6 | — | 41.7 | 80.8 | — | — | — | 11.5 | 14.7 | 80.8 | 85.3 | 57.5 | 21.6 | 82 | 36 | 19.3 | 34.5 | — | 50 | — | — | — | |
| AdaptSegBackbone=ResNet-101, Architecture=DeepLab-V2, Training Strategy=Adversarial Training2020.10 | 84.3 | — | 42.7 | 77.5 | — | — | — | 4.7 | 7 | 77.9 | 82.5 | 54.3 | 21 | 72.3 | 32.2 | 18.9 | 32.3 | — | 46.7 | — | — | — | |
| AdaptSegNetmulti-stage training=false, multi-model fusion=false2022.04 | 84.3 | — | 42.7 | 77.5 | — | — | — | 4.7 | 7 | 77.9 | 82.5 | 54.3 | 21 | 72.3 | 32.2 | 18.9 | 32.3 | — | — | 46.7 | — | — | |
| AdaptSegNetBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 84.3 | — | 42.7 | 77.5 | — | — | — | 4.7 | 7 | 77.9 | 82.5 | 54.3 | 21 | 72.3 | 32.2 | 18.9 | 32.3 | — | — | 46.7 | — | — | |
| DAP + ProDAmulti-stage training=true, multi-model fusion=false2022.04 | 84.2 | 59.8 | 46.5 | 82.5 | 35.1 | 20 | 46.7 | 53.6 | 45.7 | 89.3 | 87.5 | 75.7 | 34.6 | 91.7 | 73.5 | 49.4 | 60.5 | — | — | 64.3 | — | — | |
| DACS+DAPmulti-stage training=false, multi-model fusion=false2022.04 | 83.9 | 50.2 | 33.3 | 80.2 | 24.1 | 120 | 33.4 | 30.8 | 33.8 | 84.3 | 88.5 | 65.7 | 36.2 | 84.3 | 43.3 | 33.3 | 46.3 | — | — | 57.2 | — | — | |
| MRNet + AdaBoost StudentBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 83.6 | 45.9 | 40.1 | 81.3 | 9.6 | 0.8 | 36.8 | 25.8 | 16.3 | 84.5 | 87.1 | 60.3 | 23.9 | 84.9 | 32.3 | 19 | 48.4 | — | — | 52.9 | — | — | |
| MCDBackbone=DRN-105, k=22017.12 | 83.5 | 36.3 | 40.9 | 77.6 | 6 | 0.1 | 27.9 | 6.2 | 6 | 83.1 | 83.5 | 51.5 | 11.8 | 78.9 | 19.8 | 4.6 | 0 | — | — | — | — | — | |
| Maximum Classifier DiscrepancyBackbone=DRN-105, k=22017.12 | 83.5 | 36.3 | 40.9 | 77.6 | 6 | 0.1 | 27.9 | 6.2 | 6 | 83.1 | 83.5 | 51.5 | 11.8 | 78.9 | 19.8 | 4.6 | 0 | — | — | — | — | — | |
| PITmulti-stage training=false, multi-model fusion=false2022.04 | 83.1 | 44 | 27.6 | 81.5 | 8.9 | 30 | 21.8 | 26.4 | 33.8 | 76.4 | 78.8 | 64.2 | 27.6 | 79.6 | 31.2 | 31 | 31.3 | — | — | 51.8 | — | — | |
| PITCategory=Pure semantic segmentation based UDA2022.11 | 83.1 | 44 | 27.6 | 81.5 | 8.9 | 0.3 | 21.8 | 26.4 | 33.8 | 76.4 | 78.8 | 64.2 | 27.6 | 79.6 | 31.2 | 31 | 31.3 | — | — | — | — | — | |
| PITUDA type=Pure semantic segmentation2022.11 | 83.1 | 44 | 27.6 | 81.5 | 8.9 | 0.3 | 21.8 | 26.4 | 33.8 | 76.4 | 78.8 | 64.2 | 27.6 | 79.6 | 31.2 | 31 | 31.3 | — | — | — | — | — | |
| SIM2021.10 | 83 | — | 44 | 80.3 | — | — | — | 17.1 | 15.8 | 80.5 | 81.8 | 59.9 | 33.1 | 70.2 | 37.3 | 28.5 | 45.8 | — | 52.1 | — | — | — | |
| DTBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 83 | — | 44 | 80.3 | — | — | — | 17.1 | 15.8 | 80.5 | 81.8 | 59.9 | 33.1 | 70.2 | 37.3 | 28.5 | 45.8 | — | — | 52.1 | — | — | |
| MaxSquareArchitecture=ResNet1012020.03 | 82.9 | 48.2 | 40.7 | 80.3 | — | — | — | 12.8 | 18.2 | 82.5 | 82.2 | 53.1 | 18 | 79 | 31.4 | 10.4 | 35.6 | — | — | — | — | — | |
| MaxSquareBackbone=ResNet-101, Architecture=DeepLab-V2, Training Strategy=Self-Training2020.10 | 82.9 | — | 40.7 | 80.3 | 10.2 | 0.8 | 25.8 | 12.8 | 18.2 | 82.5 | 82.2 | 53.1 | 18 | 79 | 31.4 | 10.4 | 35.6 | 41.4 | 48.2 | — | — | — | |
| CLAN2021.10 | 82.7 | — | 37.2 | 81.5 | — | — | — | 17.1 | 13.1 | 81.2 | 83.3 | 55.5 | 22.1 | 76.6 | 30.1 | 23.5 | 30.7 | — | 48.8 | — | — | — | |
| PLCABackbone=ResNet-101, Architecture=DeepLab-V22020.10 | 82.6 | — | 29 | 81 | 11.2 | 0.2 | 33.6 | 24.9 | 18.3 | 82.8 | 82.3 | 62.1 | 26.5 | 85.6 | 48.9 | 26.8 | 52.2 | 46.8 | 54 | — | — | — | |
| SIBANArchitecture=ResNet1012020.03 | 82.5 | 46.3 | 24 | 79.4 | — | — | — | 16.5 | 12.7 | 79.2 | 82.8 | 58.3 | 18 | 79.3 | 25.3 | 17.6 | 25.9 | — | — | — | — | — | |
| CDANetArchitecture=ResNet1012020.03 | 82.5 | 52.4 | 42.2 | 81.3 | — | — | — | 18.3 | 15.9 | 80.6 | 83.5 | 61.4 | 33.2 | 72.9 | 39.3 | 26.6 | 43.9 | — | — | — | — | — | |
| SIBANBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 82.5 | — | 24 | 79.4 | — | — | — | 16.5 | 12.7 | 79.2 | 82.8 | 58.3 | 18 | 79.3 | 25.3 | 17.6 | 25.9 | — | — | 46.3 | — | — | |
| PatchAlignBackbone=ResNet-101, Architecture=DeepLab-V2, Training Strategy=Adversarial Training2020.10 | 82.4 | — | 38 | 78.6 | 8.7 | 0.6 | 26 | 3.9 | 11.1 | 75.5 | 84.6 | 53.5 | 21.6 | 71.4 | 32.6 | 19.3 | 31.7 | 40 | 46.5 | — | — | — | |
| PatchAlignmulti-stage training=false, multi-model fusion=false2022.04 | 82.4 | 40 | 38 | 78.6 | 8.7 | 60 | 26 | 3.9 | 11.1 | 75.5 | 84.6 | 53.5 | 21.6 | 71.4 | 32.6 | 19.3 | 31.7 | — | — | 46.5 | — | — | |
| PatchAlignBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 82.4 | 40 | 38 | 78.6 | 8.7 | 0.6 | 26 | 3.9 | 11.1 | 75.5 | 84.6 | 53.5 | 21.6 | 71.4 | 32.6 | 19.3 | 31.7 | — | — | 46.5 | — | — | |
| TransDA-SBackbone=Swin-S ViT, Parameters=50M2022.03 | 82.1 | 55.5 | 40.9 | 86.2 | 25.8 | 1 | 53 | 53.7 | 36.1 | 89.2 | 90.3 | 68 | 26.2 | 90.9 | 58.4 | 41.2 | 45.4 | — | — | 62.2 | — | — | |
| DACS (rep)multi-stage training=false, multi-model fusion=false, implementation=our re-implementation2022.04 | 82.1 | 48.9 | 31 | 82.4 | 22.1 | 120 | 33.1 | 32.7 | 25.1 | 84.4 | 88.2 | 65.2 | 34.3 | 83.4 | 42.9 | 24.1 | 50.8 | — | — | 55.9 | — | — | |
| MRNetBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 82 | 43.2 | 36.5 | 80.4 | 4.2 | 0.4 | 33.7 | 18 | 13.4 | 81.1 | 80.8 | 61.3 | 21.7 | 84.4 | 32.4 | 14.8 | 45.7 | — | — | 50.2 | — | — | |
| IASTmulti-stage training=false, multi-model fusion=false2022.04 | 81.9 | 49.8 | 41.5 | 83.3 | 17.7 | 460 | 32.3 | 30.9 | 28.8 | 83.4 | 85 | 65.6 | 30.8 | 86.5 | 38.2 | 33.1 | 52.7 | — | — | 57 | — | — | |
| IASTCategory=Pure semantic segmentation based UDA2022.11 | 81.9 | 49.8 | 41.5 | 83.3 | 17.7 | 4.6 | 32.3 | 30.9 | 28.8 | 83.4 | 85 | 65.5 | 30.8 | 86.5 | 38.2 | 33.1 | 52.7 | — | — | — | — | — | |
| IASTUDA type=Pure semantic segmentation2022.11 | 81.9 | 49.8 | 41.5 | 83.3 | 17.7 | 4.6 | 32.3 | 30.9 | 28.8 | 83.4 | 85 | 65.5 | 30.8 | 86.5 | 38.2 | 33.1 | 52.7 | — | — | — | — | — | |
| CLANBackbone=ResNet-101, Oracle mIoU=71.72019.04 | 81.3 | 47.8 | 37 | 80.1 | — | — | — | 16.1 | 13.7 | 78.2 | 81.5 | 53.4 | 21.2 | 73 | 32.9 | 22.6 | 30.7 | — | — | — | — | — | |
| CLANArchitecture=ResNet1012020.03 | 81.3 | 47.8 | 37 | 80.1 | — | — | — | 16.1 | 13.7 | 78.2 | 81.5 | 53.4 | 21.2 | 73 | 32.9 | 22.6 | 30.7 | — | — | — | — | — | |
| CLANBackbone=ResNet-101, Architecture=DeepLab-V2, Training Strategy=Adversarial Training2020.10 | 81.3 | — | 37 | 80.1 | — | — | — | 16.1 | 13.7 | 78.2 | 81.5 | 53.4 | 21.2 | 73 | 32.9 | 22.6 | 30.7 | — | 47.8 | — | — | — | |
| CLANBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 81.3 | — | 37 | 80.1 | — | — | — | 16.1 | 13.7 | 78.2 | 81.5 | 53.4 | 21.2 | 73 | 32.9 | 22.6 | 30.7 | — | — | 47.8 | — | — | |
| DACSmulti-stage training=false, multi-model fusion=false2022.04 | 80.6 | 48.3 | 25.1 | 81.9 | 21.5 | 290 | 37.2 | 22.7 | 24 | 83.7 | 90.8 | 67.6 | 38.3 | 82.9 | 38.9 | 28.5 | 47.6 | — | — | 54.8 | — | — | |
| DACSCategory=Pure semantic segmentation based UDA2022.11 | 80.6 | 48.3 | 25.1 | 81.9 | 21.5 | 2.9 | 37.2 | 22.7 | 24 | 83.7 | 90.8 | 67.6 | 38.3 | 82.9 | 38.9 | 28.5 | 47.6 | — | — | — | — | — | |
| DACSUDA type=Pure semantic segmentation2022.11 | 80.6 | 48.3 | 25.1 | 81.9 | 21.5 | 2.9 | 37.2 | 22.7 | 24 | 83.7 | 90.8 | 67.6 | 38.3 | 82.9 | 38.9 | 28.5 | 47.6 | — | — | — | — | — | |
| DCANBackbone=VGG-16, Oracle mIoU=59.52019.04 | 79.9 | 35.4 | 30.4 | 70.8 | 1.6 | 0.6 | 22.3 | 6.7 | — | 23 | 76.9 | 73.9 | 41.9 | 16.7 | 61.7 | 11.5 | 10.3 | — | — | — | — | — | |
| DCANArchitecture=VGG162020.03 | 79.9 | 35.4 | 30.4 | 70.8 | 1.6 | 0.6 | 22.3 | 6.7 | 23 | 76.9 | 73.9 | 41.9 | 16.7 | 61.7 | 11.5 | 10.3 | 38.6 | — | — | — | — | — | |
| CCM2021.10 | 79.6 | — | 36.4 | 80.6 | 13.3 | 0.3 | 25.5 | 22.4 | 14.9 | 81.8 | 77.4 | 56.8 | 25.9 | 80.7 | 45.3 | 29.9 | 52 | 45.2 | 52.9 | — | — | — | |
| CCMBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 79.6 | 45.2 | 36.4 | 80.6 | 13.3 | 0.3 | 25.5 | 22.4 | 14.9 | 81.8 | 77.4 | 56.8 | 25.9 | 80.7 | 45.3 | 29.9 | 52 | — | — | 52.9 | — | — | |
| FDAmulti-stage training=false, multi-model fusion=false2022.04 | 79.3 | — | 35 | 73.2 | — | — | — | 19.9 | 24 | 61.7 | 82.6 | 61.4 | 31.1 | 83.9 | 40.8 | 38.4 | 51.1 | — | — | 52.5 | — | — | |
| FDA2021.10 | 79.3 | — | 35 | 73.2 | — | — | — | 19.9 | 24 | 61.7 | 82.6 | 61.4 | 31.1 | 83.9 | 40.8 | 38.4 | 51.1 | — | 52.5 | — | — | — | |
| AdaptSegNetBackbone=ResNet-101, Oracle mIoU=71.72019.04 | 79.2 | 45.9 | 37.2 | 78.8 | — | — | — | 9.9 | 10.5 | 78.2 | 80.5 | 53.5 | 19.6 | 67 | 29.5 | 21.6 | 31.3 | — | — | — | — | — | |
| AdaptSegBackbone=ResNet-101, Parameters=43M2022.03 | 79.2 | — | 37.2 | 78.8 | — | — | — | 9.9 | 10.5 | 78.2 | 80.5 | 53.5 | 19.6 | 67 | 29.5 | 21.6 | 31.3 | — | — | 45.9 | — | — | |
| GIO-AdaArchitecture=VGG162020.03 | 78.3 | 37.3 | 29.2 | 76.9 | 11.4 | 0.3 | 26.5 | 10.8 | 17.2 | 81.7 | 81.9 | 45.8 | 15.4 | 68 | 15.9 | 7.5 | 30.4 | — | — | — | — | — | |
| GIO-AdaCategory=UDA with auxiliary tasks, Uses source depth labels=true2022.11 | 78.3 | 37.3 | 29.2 | 76.9 | 11.4 | 0.3 | 26.5 | 10.8 | 17.2 | 81.7 | 81.9 | 45.8 | 15.4 | 68 | 15.9 | 7.5 | 30.4 | — | — | — | — | — | |
| GIO-AdaUDA type=Auxiliary task based2022.11 | 78.3 | 37.3 | 29.2 | 76.9 | 11.4 | 0.3 | 26.5 | 10.8 | 17.2 | 81.7 | 81.9 | 45.8 | 15.4 | 68 | 15.9 | 7.5 | 30.4 | — | — | — | — | — | |
| MaxSquareBackbone=ResNet-101, Architecture=Deeplab-v22021.03 | 77.4 | 39.3 | 34 | 78.7 | 5.6 | 0.2 | 27.7 | 5.8 | 9.8 | 80.7 | 83.2 | 58.5 | 20.5 | 74.1 | 32.1 | 11 | 29.9 | — | — | 45.8 | — | — |