Many unsupervised domain adaptation (UDA) methods have been proposed to bridge the domain gap by utilizing domain invariant information. Most approaches have chosen depth as such information and achieved remarkable success. Despite their effectiveness, using depth as domain invariant information in UDA tasks may lead to multiple issues, such as excessively high extraction costs and difficulties in achieving a reliable prediction quality. As a result, we introduce Edge Learning based Domain Adaptation (ELDA), a framework which incorporates edge information into its training process to serve as a type of domain invariant information. In our experiments, we quantitatively and qualitatively demonstrate that the incorporation of edge information is indeed beneficial and effective and enables ELDA to outperform the contemporary state-of-the-art methods on two commonly adopted benchmarks for semantic segmentation based UDA tasks. In addition, we show that ELDA is able to better separate the feature distributions of different classes. We further provide an ablation analysis to justify our design decisions.
@article{arxiv.2211.08888,
title = {ELDA: Using Edges to Have an Edge on Semantic Segmentation Based UDA},
author = {Ting-Hsuan Liao and Huang-Ru Liao and Shan-Ya Yang and Jie-En Yao and Li-Yuan Tsao and Hsu-Shen Liu and Bo-Wun Cheng and Chen-Hao Chao and Chia-Che Chang and Yi-Chen Lo and Chun-Yi Lee},
journal= {arXiv preprint arXiv:2211.08888},
year = {2022}
}
Comments
Accepted by BMVC2022. Ting-Hsuan Liao and Huang-Ru Liao contributed equally to this work