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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Images captured in snowy days suffer from noticeable degradation of scenevisibility, which degenerates the performance of current vision-basedintelligent systems. Removing snow from images thus is an important topic incomputer vision. In this paper, we propose a Deep Dense Multi-Scale Network(\textbf{DDMSNet}) for snow removal by exploiting semantic and geometricpriors. As images captured in outdoor often share similar scenes and theirvisibility varies with depth from camera, such semantic and geometricinformation provides a strong prior for snowy image restoration. We incorporatethe semantic and geometric maps as input and learn the semantic-aware andgeometry-aware representation to remove snow. In particular, we first create acoarse network to remove snow from the input images. Then, the coarselydesnowed images are fed into another network to obtain the semantic andgeometric labels. Finally, we design a DDMSNet to learn semantic-aware andgeometry-aware representation via a self-attention mechanism to produce thefinal clean images. Experiments evaluated on public synthetic and real-worldsnowy images verify the superiority of the proposed method, offering betterresults both quantitatively and qualitatively.

Deep Dense Multi-Scale Network for Snow Removal Using Semantic and Depth Priors