Unsupervised Domain Adaptation for Semantic Segmentation by Content Transfer

Authors

  • Suhyeon Lee Yonsei University
  • Junhyuk Hyun Yonsei University
  • Hongje Seong Yonsei University
  • Euntai Kim Yonsei University

Keywords:

Transfer/Adaptation/Multi-task/Meta/Automated Learning, Unsupervised & Self-Supervised Learning

Abstract

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.

Downloads

Published

2021-05-18

How to Cite

Lee, S., Hyun, J., Seong, H., & Kim, E. (2021). Unsupervised Domain Adaptation for Semantic Segmentation by Content Transfer. Proceedings of the AAAI Conference on Artificial Intelligence, 35(9), 8306-8315. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/17010

Issue

Section

AAAI Technical Track on Machine Learning II