Exploiting Inter-pixel Correlations in Unsupervised Domain Adaptation for Semantic Segmentation

“Self-training” has become a dominant method for se-mantic segmentation via unsupervised domain adaptation (UDA). It creates a set of pseudo labels for the target do-main to give explicit supervision. However, the pseudo la-bels are noisy, sparse and do not provide any information about inter-pixel correlations. We regard inter-pixel cor-relation quite important because semantic segmentation is a task of predicting highly structured pixel-level outputs. Therefore, in this paper, we propose a method of transfer-ring the inter-pixel correlations from the source domain to the target domain via a self-attention module. The module takes the prediction of the segmentation network as an in-put and creates a self-attended prediction that correlates similar pixels. The module is trained only on the source domain to learn the domain-invariant inter-pixel correlations, then later, it is used to train the segmentation network on the target domain. The network learns not only from the pseudo labels but also by following the output of the self-attention module which provides additional knowledge about the inter-pixel correlations. Through extensive ex-periments, we show that our method significantly improves the performance on two standard UDA benchmarks and also can be combined with recent state-of-the-art method to achieve better performance.