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

Unsupervised domain adaptation (UDA) in semantic segmentation is afundamental yet promising task relieving the need for laborious annotationworks. However, the domain shifts/discrepancies problem in this task compromisethe final segmentation performance. Based on our observation, the main causesof the domain shifts are differences in imaging conditions, called image-leveldomain shifts, and differences in object category configurations calledcategory-level domain shifts. In this paper, we propose a novel UDA pipelinethat unifies image-level alignment and category-level feature distributionregularization in a coarse-to-fine manner. Specifically, on the coarse side, wepropose a photometric alignment module that aligns an image in the sourcedomain with a reference image from the target domain using a set of image-leveloperators; on the fine side, we propose a category-oriented triplet loss thatimposes a soft constraint to regularize category centers in the source domainand a self-supervised consistency regularization method in the target domain.Experimental results show that our proposed pipeline improves thegeneralization capability of the final segmentation model and significantlyoutperforms all previous state-of-the-arts.

Coarse-to-Fine Domain Adaptive Semantic Segmentation with Photometric Alignment and Category-Center Regularization