English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Many unsupervised domain adaptation (UDA) methods have been proposed tobridge the domain gap by utilizing domain invariant information. Mostapproaches have chosen depth as such information and achieved remarkablesuccess. Despite their effectiveness, using depth as domain invariantinformation in UDA tasks may lead to multiple issues, such as excessively highextraction costs and difficulties in achieving a reliable prediction quality.As a result, we introduce Edge Learning based Domain Adaptation (ELDA), aframework which incorporates edge information into its training process toserve as a type of domain invariant information. In our experiments, wequantitatively and qualitatively demonstrate that the incorporation of edgeinformation is indeed beneficial and effective and enables ELDA to outperformthe contemporary state-of-the-art methods on two commonly adopted benchmarksfor semantic segmentation based UDA tasks. In addition, we show that ELDA isable to better separate the feature distributions of different classes. Wefurther provide an ablation analysis to justify our design decisions.

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