Semantic Segmentation on Cityscapes fine (val)
80.3mIoUBiSeNet
Evaluation Results
| Method | Links | |
|---|---|---|
| BiSeNetBaseModel=Res1012018.08 | 80.3 | |
| BiSeNetBaseModel=Res182018.08 | 78.6 | |
| DUCBaseModel=Res1522018.08 | 76.7 | |
| SlotConEpochs=800, Multi crop=false, Obj. Prior=false2022.05 | 76.2 | |
| DetConEpochs=1000, Multi crop=false, Obj. Prior=true2022.05 | 76.1 | |
| Lite-HRNet-30Pretrained on ImageNet=false, Number of Parameters=1.8M, GFLOPs=3.02, Resolution=512 x 10242021.04 | 76 | |
| PC2SegBackbone=ResNet-101, Base Architecture=DeepLab-v3+2021.08 | 75.99 | |
| DenseCLEpochs=800, Multi crop=false, Obj. Prior=false2022.05 | 75.8 | |
| ORLEpochs=800, Multi crop=true, Obj. Prior=true2022.05 | 75.6 | |
| MobileNextPretrained on ImageNet=true, Number of Parameters=4.5M, GFLOPs=10.1*, Resolution=1024 x 20482021.04 | 75.5 | |
| SwiftNetPretrained on ImageNet=true, Number of Parameters=11.8M, GFLOPs=104, Resolution=1024 x 20482021.04 | 75.4 | |
| PC2SegBackbone=ResNet-50, Base Architecture=DeepLab-v3+2021.08 | 75.39 | |
| Revisit.Epochs=800, Multi crop=true, Obj. Prior=false2022.05 | 75.3 | |
| PixProEpochs=800, Multi crop=false, Obj. Prior=false2022.05 | 75.2 | |
| RELICv2Pre-training=ImageNet, Fine-tuning=true, Fine-tuning epochs=1602022.01 | 75.2 | |
| Sup. baselineBackbone=ResNet-101, Base Architecture=DeepLab-v3+2021.08 | 74.88 | |
| BiSeNetV1 BPretrained on ImageNet=true, Number of Parameters=49.0M, GFLOPs=55.3, Resolution=768 x 15362021.04 | 74.8 | |
| ShelfNetPretrained on ImageNet=true, Number of Parameters=36.9M, GFLOPs=46.8, Resolution=1024 x 20482021.04 | 74.8 | |
| BYOLPre-training=ImageNet, Fine-tuning=true, Fine-tuning epochs=1602022.01 | 74.6 | |
| Lite-HRNet-18Pretrained on ImageNet=false, Number of Parameters=1.1M, GFLOPs=1.95, Resolution=512 x 10242021.04 | 73.8 | |
| MoCo v2Epochs=800, Multi crop=false, Obj. Prior=false2022.05 | 73.8 | |
| Sup. baselineBackbone=ResNet-50, Base Architecture=DeepLab-v3+2021.08 | 73.64 | |
| BiSeNetV2 SmallPretrained on ImageNet=false, GFLOPs=21.15, Resolution=512 x 10242021.04 | 73.4 | |
| FasterSegPretrained on ImageNet=true, Number of Parameters=4.4M, GFLOPs=28.2, Resolution=1024 x 20482021.04 | 73.1 | |
| DF1-Seg-d8Pretrained on ImageNet=true, Resolution=1024 x 20482021.04 | 72.4 | |
| MobileNetV3Pretrained on ImageNet=true, Number of Parameters=1.5M, GFLOPs=9.1, Resolution=1024 x 20482021.04 | 72.4 | |
| BiSeNetBaseModel=Xception392018.08 | 72 | |
| GASPretrained on ImageNet=true, Resolution=769 x 15372021.04 | 71.8 | |
| CASPretrained on ImageNet=true, Resolution=768 x 15362021.04 | 71.6 | |
| DeepLab-v2+CRFBaseModel=Res1012018.08 | 71.4 | |
| SwiftNetPretrained on ImageNet=true, Number of Parameters=11.8M, GFLOPs=26, Resolution=512 x 10242021.04 | 70.2 | |
| LRRBaseModel=VGG162018.08 | 70 | |
| BiSeNetV1 APretrained on ImageNet=true, Number of Parameters=5.8M, GFLOPs=14.8, Resolution=768 x 15362021.04 | 69 | |
| Dilation10BaseModel=VGG162018.08 | 68.7 | |
| Fast-SCNNPretrained on ImageNet=false, Resolution=1024 x 20482021.04 | 68.6 | |
| MobileNet V2 0.5Pretrained on ImageNet=true, Number of Parameters=0.3M, GFLOPs=3.73, Resolution=512 x 10242021.04 | 68.6 | |
| HRNet-W16Pretrained on ImageNet=true, Number of Parameters=2.0M, GFLOPs=7.8, Resolution=512 x 10242021.04 | 68.6 | |
| MobileNet V3-SmallPretrained on ImageNet=true, Number of Parameters=0.5M, GFLOPs=2.7, Resolution=512 x 10242021.04 | 68.4 | |
| DMTBackbone=ResNet-101, Base Architecture=DeepLab-v22021.08 | 68.16 | |
| ICNetPretrained on ImageNet=true, GFLOPs=28.3, Resolution=1024 x 20482021.04 | 67.7 | |
| sGANBackbone=ResNet-101, Base Architecture=DeepLab-v22021.08 | 65.8 | |
| random init.Multi crop=false, Obj. Prior=false2022.05 | 65.3 |