Unsupervised Domain Adaptation for Semantic Segmentation by Content Transfer
This work provides an incremental improvement for researchers working on unsupervised domain adaptation for semantic segmentation.
This paper addresses unsupervised domain adaptation for semantic segmentation, aiming to segment unlabeled real data using labeled synthetic data. The authors tackle the domain gap by separating image content and style, proposing a zero-style loss, and mitigate class imbalance by transferring tail class content from synthetic to real domains. The method achieves state-of-the-art performance on two major UDA settings.
In this paper, we tackle the unsupervised domain adaptation (UDA) for semantic segmentation, which aims to segment the unlabeled real data using labeled synthetic data. The main problem of UDA for semantic segmentation relies on reducing the domain gap between the real image and synthetic image. To solve this problem, we focused on separating information in an image into content and style. Here, only the content has cues for semantic segmentation, and the style makes the domain gap. Thus, precise separation of content and style in an image leads to effect as supervision of real data even when learning with synthetic data. To make the best of this effect, we propose a zero-style loss. Even though we perfectly extract content for semantic segmentation in the real domain, another main challenge, the class imbalance problem, still exists in UDA for semantic segmentation. We address this problem by transferring the contents of tail classes from synthetic to real domain. Experimental results show that the proposed method achieves the state-of-the-art performance in semantic segmentation on the major two UDA settings.