Semantic Segmentation on Cross-City Taipei
86Road IoUFADA
Evaluation results
| Method | Links | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| FADABackbone=DeepLab-v22020.07 | 86 | 42.3 | 86.1 | 16.2 | 20.5 | 78.3 | 92.7 | 47.2 | 17.7 | 72.2 | 37.2 | 54.3 | 44 | 52.7 | |
| BaselineBackbone=ResNet-101, Adaptation Level=None2018.02 | 83.5 | 33.4 | 86.6 | 12.7 | 16.4 | 77 | 92.1 | 17.6 | 13.7 | 70.7 | 37.7 | 44.4 | 18.5 | 46.5 | |
| AdaptSegNetBackbone=ResNet-101, Adaptation Level=feature space2018.02 | 82.1 | 31.9 | 84.1 | 25.7 | 13.2 | 77.2 | 81.2 | 28.1 | 12 | 67 | 35.8 | 43.5 | 20.9 | 46.6 | |
| AdaptSegNetBackbone=ResNet-101, Adaptation Level=output space2018.02 | 81.7 | 29.5 | 85.2 | 26.4 | 15.6 | 76.7 | 91.7 | 31 | 12.5 | 71.5 | 41.1 | 47.3 | 27.7 | 49.1 | |
| AdaptSegNetBackbone=DeepLab-v22020.07 | 81.7 | 29.5 | 85.2 | 26.4 | 10.6 | 76.7 | 91.7 | 31 | 12.5 | 71.5 | 41.1 | 47.3 | 27.7 | 49.1 | |
| Cross-City2018.02 | 78.6 | 28.6 | 80 | 13.1 | 7.6 | 68.2 | 82.1 | 16.8 | 9.4 | 60.4 | 34 | 26.5 | 9.9 | 39.6 | |
| Cross-CityBackbone=Dilation-Frontend2020.07 | 78.6 | 28.6 | 80 | 13.1 | 7.6 | 68.2 | 82.1 | 16.8 | 9.4 | 60.4 | 34 | 26.5 | 9.9 | 39.6 | |
| Source DeepLab-v2Backbone=DeepLab-v2, Adaptation=None2020.07 | 78.6 | 28.6 | 80 | 13.1 | 7.6 | 68.2 | 82.1 | 16.8 | 9.4 | 60.4 | 34 | 26.5 | 9.9 | 39.6 | |
| Source Dilation-FrontendBackbone=Dilation-Frontend, Adaptation=None2020.07 | 77.2 | 20.9 | 76 | 4.4 | 5.9 | 60.3 | 81.4 | 10.9 | 11 | 54.9 | 32.6 | 15.3 | 5.2 | 35.1 |