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

While nowadays deep neural networks achieve impressive performances onsemantic segmentation tasks, they are usually trained by optimizing pixel-wiselosses such as cross-entropy. As a result, the predictions outputted by suchnetworks usually struggle to accurately capture the object boundaries andexhibit holes inside the objects. In this paper, we propose a novel approach toimprove the structure of the predicted segmentation masks. We introduce a novelsemantic edge detection network, which allows to match the predicted and groundtruth segmentation masks. This Semantic Edge-Aware strategy (SEMEDA) can becombined with any backbone deep network in an end-to-end training framework.Through thorough experimental validation on Pascal VOC 2012 and Cityscapesdatasets, we show that the proposed SEMEDA approach enhances the structure ofthe predicted segmentation masks by enforcing sharp boundaries and avoidingdiscontinuities inside objects, improving the segmentation performance. Inaddition, our semantic edge-aware loss can be integrated into any popularsegmentation network without requiring any additional annotation and withnegligible computational load, as compared to standard pixel-wise cross-entropyloss.

SEMEDA: Enhancing Segmentation Precision with Semantic Edge Aware Loss