A Simple Baseline for Semi-supervised Semantic Segmentation with Strong Data Augmentation

Open paper
Benchmark Model Rank Results
semi-supervised-semantic-segmentation-on-cityscapes-12-5-labeledSimpleBaseline(DeeplabV3+ with ImageNet pretrained Xception65, sinle scale inference)#18Validation mIoU: 74.1%
semi-supervised-semantic-segmentation-on-cityscapes-25-labeledSimpleBaseline(DeepLabv3+ with ImageNet pretrained Xception65, single scale inference)#17Validation mIoU: 77.8%
semi-supervised-semantic-segmentation-on-cityscapes-50-labeledSimpleBaseline(DeepLabv3+ with ImageNet pretrained Xception65, single scale inference)#16Validation mIoU: 78.7%

Rank counts only results with a code link.