Semi Supervised Semantic Segmentation

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

Best model: GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)

Best model: GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)

Best model: S4MC

Best model: GuidedMix-Net(DeepLab v2 with ResNet50, ImageNet pretrained)

Best model: GuidedMix-Net(DeepLab v2 with ResNet50, ImageNet pretrained)

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

Best model: MMS (20% Labeled)

Best model: LaserMix (Voxel)

Best model: PLE (Voxel)

Best model: SemiVL (ViT-B/16)

Best model: M3L (Linear Fusion B2)

Best model: SemiSegContrast (DeepLab v3+ with ResNet-50 backbone, MSCOCO pretrained)

Best model: SemiVL (ViT-B/16)

Best model: MMS(20% labeled)

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

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

Best model: AllSpark

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

Best model: UniMatch V2

Best model: IM++ (416x208, 2.7m parameters, no pretraining)

Best model: AllSpark

Best model: FishSegSSL

Best model: ReCo (DeepLab v3+ with ResNet-101 backbone, ImageNet pretrained)

Best model: ReCo (DeepLab v3+ with ResNet-101 backbone, ImageNet pretrained)

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

Best model: ClassMix (DeepLab v2 MSCOCO pretrained)

Best model: Dual Teacher

Best model: SemiVL