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

Shadows often occur when we capture the documents with casual equipment,which influences the visual quality and readability of the digital copies.Different from the algorithms for natural shadow removal, the algorithms indocument shadow removal need to preserve the details of fonts and figures inhigh-resolution input. Previous works ignore this problem and remove theshadows via approximate attention and small datasets, which might not work inreal-world situations. We handle high-resolution document shadow removaldirectly via a larger-scale real-world dataset and a carefully designedfrequency-aware network. As for the dataset, we acquire over 7k couples ofhigh-resolution (2462 x 3699) images of real-world document pairs with varioussamples under different lighting circumstances, which is 10 times larger thanexisting datasets. As for the design of the network, we decouple thehigh-resolution images in the frequency domain, where the low-frequency detailsand high-frequency boundaries can be effectively learned via the carefullydesigned network structure. Powered by our network and dataset, the proposedmethod clearly shows a better performance than previous methods in terms ofvisual quality and numerical results. The code, models, and dataset areavailable at: https://github.com/CXH-Research/DocShadow-SD7K

High-Resolution Document Shadow Removal via A Large-Scale Real-World Dataset and A Frequency-Aware Shadow Erasing Net