G2L: A Global to Local Alignment Method for Unsupervised Domain Adaptive Semantic Segmentation

Benchmark Model Rank Results
domain-adaptation-on-gta5-to-cityscapesG2LmIoU: 59.7
image-to-image-translation-on-gtav-toG2LmIoU: 59.7
semantic-segmentation-on-gtav-to-cityscapes-1G2LmIoU: 59.7
synthetic-to-real-translation-on-gtav-toG2LmIoU: 59.7
synthetic-to-real-translation-on-synthia-to-1G2LMIoU (16 classes): 56.8MIoU (13 classes): 64.4
unsupervised-domain-adaptation-on-gtav-toG2LmIoU: 59.7
unsupervised-domain-adaptation-on-synthia-toG2LmIoU (13 classes): 64.4mIoU: 56.8