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

Document structure extraction has been a widely researched area for decadeswith recent works performing it as a semantic segmentation task over documentimages using fully-convolution networks. Such methods are limited by imageresolution due to which they fail to disambiguate structures in dense regionswhich appear commonly in forms. To mitigate this, we propose Form2Seq, a novelsequence-to-sequence (Seq2Seq) inspired framework for structure extractionusing text, with a specific focus on forms, which leverages relative spatialarrangement of structures. We discuss two tasks; 1) Classification of low-levelconstituent elements (TextBlock and empty fillable Widget) into ten types suchas field captions, list items, and others; 2) Grouping lower-level elementsinto higher-order constructs, such as Text Fields, ChoiceFields andChoiceGroups, used as information collection mechanism in forms. To achievethis, we arrange the constituent elements linearly in natural reading order,feed their spatial and textual representations to Seq2Seq framework, whichsequentially outputs prediction of each element depending on the final task. Wemodify Seq2Seq for grouping task and discuss improvements obtained throughcascaded end-to-end training of two tasks versus training in isolation.Experimental results show the effectiveness of our text-based approachachieving an accuracy of 90% on classification task and an F1 of 75.82, 86.01,61.63 on groups discussed above respectively, outperforming segmentationbaselines. Further we show our framework achieves state of the results fortable structure recognition on ICDAR 2013 dataset.

Form2Seq : A Framework for Higher-Order Form Structure Extraction