Semi Supervised Semantic Segmentation

Bestes Modell: UniMatch (DeepLab v3+ with ImageNet-pretrained ResNet-101, single scale inference)

Bestes Modell: GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)

Bestes Modell: GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)

Bestes Modell: S4MC

Bestes Modell: GuidedMix-Net(DeepLab v2 with ResNet50, ImageNet pretrained)

Bestes Modell: GuidedMix-Net(DeepLab v2 with ResNet50, ImageNet pretrained)

Bestes Modell: CPS (DeepLab v3+ with ImageNet-pretrained ResNet-101, single scale inference)

Bestes Modell: MMS (20% Labeled)

Bestes Modell: LaserMix (Voxel)

Bestes Modell: PLE (Voxel)

Bestes Modell: SemiVL (ViT-B/16)

Bestes Modell: M3L (Linear Fusion B2)

Bestes Modell: SemiSegContrast (DeepLab v3+ with ResNet-50 backbone, MSCOCO pretrained)

Bestes Modell: SemiVL (ViT-B/16)

Bestes Modell: MMS(20% labeled)

Bestes Modell: AEL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)

Bestes Modell: AEL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)

Bestes Modell: AllSpark

Bestes Modell: Dense FixMatch (DeepLabv3+ ResNet-101, over-sampling, single pass eval)

Bestes Modell: UniMatch V2

Bestes Modell: IM++ (416x208, 2.7m parameters, no pretraining)

Bestes Modell: AllSpark

Bestes Modell: FishSegSSL

Bestes Modell: ReCo (DeepLab v3+ with ResNet-101 backbone, ImageNet pretrained)

Bestes Modell: ReCo (DeepLab v3+ with ResNet-101 backbone, ImageNet pretrained)

Bestes Modell: ReCo (DeepLab v3+ with ResNet-101 backbone, ImageNet pre-trained)

Bestes Modell: ClassMix (DeepLab v2 MSCOCO pretrained)

Bestes Modell: Dual Teacher

Bestes Modell: SemiVL