Switching Temporary Teachers for Semi-Supervised Semantic Segmentation

Benchmark Model Rank Results
semi-supervised-semantic-segmentation-on-1Dual Teacher (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)#6Validation mIoU: 79.46
semi-supervised-semantic-segmentation-on-2Dual Teacher#6Validation mIoU: 78.4
semi-supervised-semantic-segmentation-on-21Dual Teacher#8Validation mIoU: 78.82
semi-supervised-semantic-segmentation-on-22Dual Teacher (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)#7Validation mIoU: 76.81
semi-supervised-semantic-segmentation-on-4Dual Teacher#7Validation mIoU: 81.19
semi-supervised-semantic-segmentation-on-8Dual Teacher#4Validation mIoU: 80.52
semi-supervised-semantic-segmentation-on-9Dual Teacher#2Validation mIoU: 81.03