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

Adversarial learning-based image defogging methods have been extensivelystudied in computer vision due to their remarkable performance. However, mostexisting methods have limited defogging capabilities for real cases becausethey are trained on the paired clear and synthesized foggy images of the samescenes. In addition, they have limitations in preserving vivid color and richtextual details in defogging. To address these issues, we develop a novelgenerative adversarial network, called quad-path cycle consistent adversarialnetwork (QPC-Net), for single image defogging. QPC-Net consists of aFog2Fogfree block and a Fogfree2Fog block. In each block, there are threelearning-based modules, namely, fog removal, color-texture recovery, and fogsynthetic, which sequentially compose dual-path that constrain each other togenerate high quality images. Specifically, the color-texture recovery model isdesigned to exploit the self-similarity of texture and structure information bylearning the holistic channel-spatial feature correlations between the foggyimage with its several derived images. Moreover, in the fog synthetic module,we utilize the atmospheric scattering model to guide it to improve thegenerative quality by focusing on an atmospheric light optimization with anovel sky segmentation network. Extensive experiments on both synthetic andreal-world datasets show that QPC-Net outperforms state-of-the-art defoggingmethods in terms of quantitative accuracy and subjective visual quality.

Unpaired Quad-Path Cycle Consistent Adversarial Networks for Single Image Defogging