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/

Zendy Plus

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

Zendy Tools

Zendy Open

Structure extraction from document images has been a long-standing researchtopic due to its high impact on a wide range of practical applications. In thispaper, we share our findings on employing a hierarchical semantic segmentationnetwork for this task of structure extraction. We propose a prior based deephierarchical CNN network architecture that enables document structureextraction using very high resolution(1800 x 1000) images. We divide thedocument image into overlapping horizontal strips such that the networksegments a strip and uses its prediction mask as prior for predicting thesegmentation of the subsequent strip. We perform experiments establishing theeffectiveness of our strip based network architecture through ablation methodsand comparison with low-resolution variations. Further, to demonstrate ournetwork's capabilities, we train it on only one type of documents (Forms) andachieve state-of-the-art results over other general document datasets. Weintroduce our new human-annotated forms dataset and show that our methodsignificantly outperforms different segmentation baselines on this dataset inextracting hierarchical structures. Our method is currently being used inAdobe's AEM Forms for automated conversion of paper and PDF forms to modernHTML based forms.

Document Structure Extraction using Prior based High Resolution Hierarchical Semantic Segmentation