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

In the real world, the degradation of images taken under haze can be quitecomplex, where the spatial distribution of haze is varied from image to image.Recent methods adopt deep neural networks to recover clean scenes from hazyimages directly. However, due to the paradox caused by the variation of realcaptured haze and the fixed degradation parameters of the current networks, thegeneralization ability of recent dehazing methods on real-world hazy images isnot ideal.To address the problem of modeling real-world haze degradation, wepropose to solve this problem by perceiving and modeling density for unevenhaze distribution. We propose a novel Separable Hybrid Attention (SHA) moduleto encode haze density by capturing features in the orthogonal directions toachieve this goal. Moreover, a density map is proposed to model the unevendistribution of the haze explicitly. The density map generates positionalencoding in a semi-supervised way. Such a haze density perceiving and modelingcapture the unevenly distributed degeneration at the feature level effectively.Through a suitable combination of SHA and density map, we design a noveldehazing network architecture, which achieves a good complexity-performancetrade-off. The extensive experiments on two large-scale datasets demonstratethat our method surpasses all state-of-the-art approaches by a large marginboth quantitatively and qualitatively, boosting the best published PSNR metricfrom 28.53 dB to 33.49 dB on the Haze4k test dataset and from 37.17 dB to 38.41dB on the SOTS indoor test dataset.

Perceiving and Modeling Density is All You Need for Image Dehazing