[ Supplementary Material ] Learning to Adapt via Latent Domains for Adaptive Semantic Segmentation Anonymous Author(s) Affiliation Address email A Appendix 1

In A.1, we use ResNet101 as the backbone network and compare our method with state-of-3 In A.2, we provide more t-SNE visualization results for a comprehensive analysis on the In A.3, we study the effect of the image-to-image translation model on the performance of In A.4, we discuss the limitations of our method and provide the URL link of code to " indicate the method using VGG16 and ResNet101 We also provide some qualitative semantic segmentation results in Figure. 1, where we observe To reproduce our main experimental results, we release the code at: Code link. PyTorch=1.2.0 installed following the official instructions (https://pytorch.org)