Contrastive Learning and Self-Training for Unsupervised Domain Adaptation in Semantic Segmentation

Open paper
Benchmark Model Rank Results
synthetic-to-real-translation-on-gtav-toCLST–mIoU: 51.6
synthetic-to-real-translation-on-synthia-to-1CLST(ResNet-101)–MIoU (16 classes): 49.8MIoU (13 classes): 57.8

Rank counts only results with a code link.