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

Text detection and recognition are essential components of a modern OCRsystem. Most OCR approaches attempt to obtain accurate bounding boxes of textat the detection stage, which is used as the input of the text recognitionstage. We observe that when using tight text bounding boxes as input, a textrecognizer frequently fails to achieve optimal performance due to theinconsistency between bounding boxes and deep representations of textrecognition. In this paper, we propose Box Adjuster, a reinforcementlearning-based method for adjusting the shape of each text bounding box to makeit more compatible with text recognition models. Additionally, when dealingwith cross-domain problems such as synthetic-to-real, the proposed methodsignificantly reduces mismatches in domain distribution between the source andtarget domains. Experiments demonstrate that the performance of end-to-end textrecognition systems can be improved when using the adjusted bounding boxes asthe ground truths for training. Specifically, on several benchmark datasets forscene text understanding, the proposed method outperforms state-of-the-art textspotters by an average of 2.0% F-Score on end-to-end text recognition tasks and4.6% F-Score on domain adaptation tasks.

Optimal Boxes: Boosting End-to-End Scene Text Recognition by Adjusting Annotated Bounding Boxes via Reinforcement Learning