Smoothing Matters: Momentum Transformer for Domain Adaptive Semantic Segmentation

Benchmark Model Rank Results
domain-adaptation-on-gta5-to-cityscapesTransDA-B#13mIoU: 63.9
image-to-image-translation-on-gtav-toTransDA-B#9mIoU: 63.9
image-to-image-translation-on-synthia-toTransDA-B#8mIoU (13 classes): 66.3
semantic-segmentation-on-gtav-to-cityscapes-1TransDA-B#7mIoU: 63.9
semantic-segmentation-on-synthia-toTransDA-B#7Mean IoU: 59.3
synthetic-to-real-translation-on-gtav-toTransDA-B#13mIoU: 63.9
synthetic-to-real-translation-on-synthia-to-1TransDA-B#11MIoU (16 classes): 59.3MIoU (13 classes): 66.3
unsupervised-domain-adaptation-on-gtav-toTransDA-B#11mIoU: 63.9
unsupervised-domain-adaptation-on-synthia-toTransDA-B#12mIoU (13 classes): 66.3