ELDA: Using Edges to Have an Edge on Semantic Segmentation Based UDA

TL;DR

ELDA introduces edge information into unsupervised domain adaptation for semantic segmentation, offering a cost-effective alternative to depth that improves performance and class separation.

Contribution

The paper proposes a novel framework, ELDA, that leverages edge information as a domain invariant feature for UDA, outperforming existing methods.

Findings

ELDA outperforms state-of-the-art methods on benchmark datasets.

Edge information improves class separation in feature space.

ELDA is more cost-effective than depth-based approaches.

Abstract

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.