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

Scene text recognition is a popular topic and extensively used in theindustry. Although many methods have achieved satisfactory performance for theclose-set text recognition challenges, these methods lose feasibility inopen-set scenarios, where collecting data or retraining models for novelcharacters could yield a high cost. For example, annotating samples for foreignlanguages can be expensive, whereas retraining the model each time when a novelcharacter is discovered from historical documents costs both time andresources. In this paper, we introduce and formulate a new open-set textrecognition task which demands the capability to spot and recognize novelcharacters without retraining. A label-to-prototype learning framework is alsoproposed as a baseline for the proposed task. Specifically, the frameworkintroduces a generalizable label-to-prototype mapping function to buildprototypes (class centers) for both seen and unseen classes. An open-setpredictor is then utilized to recognize or reject samples according to theprototypes. The implementation of rejection capability over out-of-setcharacters allows automatic spotting of unknown characters in the incoming datastream. Extensive experiments show that our method achieves promisingperformance on a variety of zero-shot, close-set, and open-set text recognitiondatasets

Towards Open-Set Text Recognition via Label-to-Prototype Learning