Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels

Benchmark Model Rank Results
semi-supervised-semantic-segmentation-on-1U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, AEL)#12Validation mIoU: 78.51%
semi-supervised-semantic-segmentation-on-10U2PL (DeepLab v3+ with ResNet-101)#12Validation mIoU: 79.5
semi-supervised-semantic-segmentation-on-15U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, CutMix)#3Validation mIoU: 80.5%
semi-supervised-semantic-segmentation-on-2U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, AEL)#15Validation mIoU: 76.48%
semi-supervised-semantic-segmentation-on-21U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, CutMix)#10Validation mIoU: 77.21
semi-supervised-semantic-segmentation-on-22U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, AEL)#12Validation mIoU: 74.90%
semi-supervised-semantic-segmentation-on-27U2PL (DeepLab v3+ with ResNet-101)#14Validation mIoU: 68.0
semi-supervised-semantic-segmentation-on-28U2PL (DeepLab v3+ with ResNet-101)#13Validation mIoU: 69.2
semi-supervised-semantic-segmentation-on-29U2PL (DeepLab v3+ with ResNet-101)#12Validation mIoU: 73.7
semi-supervised-semantic-segmentation-on-30U2PL (DeepLab v3+ with ResNet-101)#13Validation mIoU: 76.2
semi-supervised-semantic-segmentation-on-4U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, CutMix)#10Validation mIoU: 79.01%
semi-supervised-semantic-segmentation-on-8U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, AEL)#13Validation mIoU: 79.12%
semi-supervised-semantic-segmentation-on-9U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, CutMix)#10Validation mIoU: 79.3