Semantic Segmentation on GTA5
60.7mIoUCMFormer
Evaluation Results
| Method | Links | |
|---|---|---|
| CMFormerBackbone=Swin-L, Trained on Cityscapes=true2023.07 | 60.7 | |
| HGFormerBackbone=Swin-L, Trained on Cityscapes=true2023.07 | 59.4 | |
| CMFormerBackbone=Swin-B, Trained on Cityscapes=true2023.07 | 58.11 | |
| Mask2FormerBackbone=Swin-L, Trained on Cityscapes=true2023.07 | 57.8 | |
| Mask2FormerBackbone=Swin-B, Trained on Cityscapes=true2023.07 | 55.05 | |
| HGFormerBackbone=Swin-T, Trained on Cityscapes=true2023.07 | 51.3 | |
| Mask2FormerBackbone=Swin-T, Trained on Cityscapes=true2023.07 | 50.6 | |
| SHADEBackbone=ResNet50, Trained on Cityscapes=true2023.07 | 48.61 | |
| SAWBackbone=ResNet50, Trained on Cityscapes=true2023.07 | 47.28 | |
| WildNetBackbone=ResNet50, Trained on Cityscapes=true2023.07 | 47.01 | |
| DIRLBackbone=ResNet50, Trained on Cityscapes=true2023.07 | 46.52 | |
| GTRBackbone=ResNet50, Trained on Cityscapes=true2023.07 | 45.79 | |
| IternormBackbone=ResNet50, Trained on Cityscapes=true2023.07 | 45.73 | |
| DRPCBackbone=ResNet50, Trained on Cityscapes=true2023.07 | 45.62 | |
| IBNBackbone=ResNet50, Trained on Cityscapes=true2023.07 | 45.06 | |
| ISWBackbone=ResNet50, Trained on Cityscapes=true2023.07 | 45 | |
| IWBackbone=ResNet50, Trained on Cityscapes=true2023.07 | 44.87 | |
| ULDASource Domain=Cityscapes, Prompt=driving in a game, Backbone=CLIP-ResNet-50, Base Model=DeepLabv3+2024.04 | 42.91 | |
| PØDAAdaptation setting=0-shot, Prompt=driving in a game2024.10 | 41.07 | |
| PØDASource domain=Cityscapes (CS), Target prompt (P)=driving in a game, Zero-shot domain adaptation=true2024.10 | 41.07 | |
| PIDAAdaptation setting=1-shot, Target image usage=single target image2024.10 | 41.02 | |
| PØDASource Domain=Cityscapes, Prompt=driving in a game, Backbone=CLIP-ResNet-50, Base Model=DeepLabv3+2024.04 | 40.77 | |
| CMFormerTrained on=SYNTHIA (S)2023.07 | 40.65 | |
| AdaINAdaptation setting=1-shot, Target image usage=single target image2024.10 | 40.56 | |
| source-onlySource Domain=Cityscapes, Prompt=driving in a game, Backbone=CLIP-ResNet-50, Base Model=DeepLabv3+2024.04 | 39.59 | |
| source-only2024.10 | 39.59 | |
| source-onlySource domain=Cityscapes (CS), Target prompt (P)=driving in a game, Zero-shot domain adaptation=true2024.10 | 39.59 | |
| CLIPstylerSource Domain=Cityscapes, Prompt=driving in a game, Backbone=CLIP-ResNet-50, Base Model=DeepLabv3+2024.04 | 38.73 | |
| CLIPstylerSource domain=Cityscapes (CS), Target prompt (P)=driving in a game, Zero-shot domain adaptation=true2024.10 | 38.73 | |
| SAWTrained on=SYNTHIA (S)2023.07 | 29.16 | |
| DRPCTrained on=SYNTHIA (S)2023.07 | 28.75 | |
| GTRTrained on=SYNTHIA (S)2023.07 | 28.02 | |
| ISWTrained on=SYNTHIA (S)2023.07 | 27.68 | |
| IBNTrained on=SYNTHIA (S)2023.07 | 26.9 | |
| IWTrained on=SYNTHIA (S)2023.07 | 26.51 |