Semantic Segmentation on GTA5 to C-driving (test)
27.1mIoU (Rainy)DHA
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| DHABackbone=VGG-16, Training scheme=Longer (150K iterations)2021.10 | 27.1 | 30.4 | 35.5 | 36.1 | 32 | 32.3 | |
| DHABackbone=VGG-16, Training scheme=Short (5K iterations)2021.10 | 27 | 26.3 | 30.7 | 32.8 | 28.5 | 29.2 | |
| DHABackbone=VGG-16, Architecture=DeepLab V22021.10 | 25.1 | 27.6 | 33.2 | 32.6 | 29.2 | 29.6 | |
| AdventBackbone=VGG-16, Architecture=DeepLab V22021.10 | 24.6 | 26.8 | 30.9 | 31 | 28 | 28.3 | |
| AdaptSegBackbone=VGG-16, Architecture=DeepLab V22021.10 | 24.2 | 26.2 | 31.6 | 31.2 | 27.9 | 28.3 | |
| Source OnlyBackbone=VGG-16, Training scheme=Longer (150K iterations)2021.10 | 23.3 | 24 | 28.2 | 30.2 | 25.7 | 26.4 | |
| CBSTBackbone=VGG-16, Architecture=DeepLab V22021.10 | 23.1 | 25.1 | 30.1 | 30 | 26.5 | 27 | |
| CRSTBackbone=VGG-16, Architecture=DeepLab V22021.10 | 23.1 | 25.1 | 30.1 | 30.1 | 26.6 | 27.1 | |
| Source OnlyBackbone=VGG-16, Architecture=DeepLab V22021.10 | 22.8 | 24.6 | 29 | 29.5 | 25.9 | 26.5 | |
| Liu et.al.Backbone=VGG-16, Training scheme=Short (5K iterations)2021.10 | 22 | 22.9 | 27 | 27.9 | 24.5 | 25 | |
| PyCDABackbone=VGG-16, Training scheme=Short (5K iterations)2021.10 | 21.7 | 22.3 | 25.9 | 25.4 | 23.3 | 23.8 | |
| CBSTBackbone=VGG-16, Training scheme=Short (5K iterations)2021.10 | 21.3 | 20.6 | 23.9 | 24.7 | 22.2 | 22.6 | |
| IBN-NetBackbone=VGG-16, Training scheme=Short (5K iterations)2021.10 | 20.6 | 21.9 | 26.1 | 25.5 | 22.8 | 23.5 | |
| AdaptSegBackbone=VGG-16, Training scheme=Short (5K iterations)2021.10 | 20.2 | 21.2 | 23.8 | 25.1 | 22.1 | 22.5 | |
| Source OnlyBackbone=VGG-16, Training scheme=Short (5K iterations)2021.10 | 16.2 | 18 | 20.9 | 21.2 | 18.9 | 19.1 |