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

Although Convolutional Neural Networks (CNN) have made good progress in imagerestoration, the intrinsic equivalence and locality of convolutions stillconstrain further improvements in image quality. Recent vision transformer andself-attention have achieved promising results on various computer visiontasks. However, directly utilizing Transformer for image restoration is achallenging task. In this paper, we introduce an effective hybrid architecturefor sand image restoration tasks, which leverages local features from CNN andlong-range dependencies captured by transformer to improve the results further.We propose an efficient hybrid structure for sand dust image restoration tosolve the feature inconsistency issue between Transformer and CNN. Theframework complements each representation by modulating features from theCNN-based and Transformer-based branches rather than simply adding orconcatenating features. Experiments demonstrate that SandFormer achievessignificant performance improvements in synthetic and real dust scenes comparedto previous sand image restoration methods.

SANDFORMER: CNN and Transformer under Gated Fusion for Sand Dust Image Restoration