Semi-Supervised Semantic Segmentation with High- and Low-level Consistency

Benchmark Model Rank Results
semi-supervised-semantic-segmentation-on-cityscapes-12-5-labeleds4GAN (DeepLab v2 ImageNet pre-trained)#30Validation mIoU: 59.3%
semi-supervised-semantic-segmentation-on-cityscapes-25-labeleds4GAN (DeepLab v2 ImageNet pre-trained)#27Validation mIoU: 61.9%
semi-supervised-semantic-segmentation-on-pascal-voc-2012-12-5-labeleds4GAN + MLMT#27Validation mIoU: 71.4%
semi-supervised-semantic-segmentation-on-pascal-voc-2012-12-5-labeleds4GAN+MLMT#30Validation mIoU: 70.4%
semi-supervised-semantic-segmentation-on-pascal-voc-2012-12-5-labeleds4GAN+MLMT#32Validation mIoU: 67.3%
semi-supervised-semantic-segmentation-on-pascal-voc-2012-2-labeleds4GAN + MLMT (DeepLab v2 MSCOCO/ImageNet pre-trained)#8Validation mIoU: 63.3%
semi-supervised-semantic-segmentation-on-pascal-voc-2012-2-labeleds4GAN+MLMT (DeepLab v3+ ImageNet pre-trained)#9Validation mIoU: 62.6%
semi-supervised-semantic-segmentation-on-pascal-voc-2012-2-labeleds4GAN+MLMT (DeepLab v2 ImageNet pre-trained)#10Validation mIoU: 60.4%
semi-supervised-semantic-segmentation-on-pascal-voc-2012-5-labeleds4GAN + MLMT (DeepLab v2 MSCOCO/ImageNet pre-trained)#9Validation mIoU: 67.2%
semi-supervised-semantic-segmentation-on-pascal-voc-2012-5-labeleds4GAN+MLMT (DeepLab v3+ ImageNet pre-trained)#10Validation mIoU: 66.6%
semi-supervised-semantic-segmentation-on-pascal-voc-2012-5-labeleds4GAN+MLMT (DeepLab v2 ImageNet pre-trained)#12Validation mIoU: 62.9%