Since annotating pixel-level labels for semantic segmentation is laborious, leveraging synthetic data is an attractive solution. However, due to the domain gap between synthetic domain and real domain, it is challenging for a model trained with synthetic data to generalize to real data. In this paper, considering the fundamental difference between the two domains as the texture, we propose a method to adapt to the texture of the target domain. First, we diversity the texture of synthetic images using a style transfer algorithm. The various textures of generated images prevent a segmentation model from overfitting to one specific (synthetic) texture. Then, we fine-tune the model with self-training to get direct supervision of the target texture. Our results achieve state-of-the-art performance and we analyze the properties of the model trained on the stylized dataset with extensive experiments.
@article{arxiv.2003.00867,
title = {Learning Texture Invariant Representation for Domain Adaptation of Semantic Segmentation},
author = {Myeongjin Kim and Hyeran Byun},
journal= {arXiv preprint arXiv:2003.00867},
year = {2020}
}