Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386907174> ?p ?o ?g. }
Showing items 1 to 49 of
49
with 100 items per page.
- W4386907174 abstract "Offline handwriting recognition (HWR) has improved significantly with the advent of deep learning architectures in recent years. Nevertheless, it remains a challenging problem and practical applications often rely on post-processing techniques for restricting the predicted words via lexicons or language models. Despite their enhanced performance, such systems are less usable in contexts where out-of-vocabulary words are anticipated, e.g. for detecting misspelled words in school assessments. To that end, we introduce the task of comparing a handwriting image to text. To solve the problem, we propose an unrestricted binary classifier, consisting of a HWR feature extractor and a multimodal classification head which convolves the feature extractor output with the vector representation of the input text. Our model's classification head is trained entirely on synthetic data created using a state-of-the-art generative adversarial network. We demonstrate that, while maintaining high recall, the classifier can be calibrated to achieve an average precision increase of 19.5% compared to addressing the task by directly using state-of-the-art HWR models. Such massive performance gains can lead to significant productivity increases in applications utilizing human-in-the-loop automation." @default.
- W4386907174 created "2023-09-21" @default.
- W4386907174 creator A5006952036 @default.
- W4386907174 creator A5080429518 @default.
- W4386907174 date "2023-09-18" @default.
- W4386907174 modified "2023-10-16" @default.
- W4386907174 title "Offline Detection of Misspelled Handwritten Words by Convolving Recognition Model Features with Text Labels" @default.
- W4386907174 doi "https://doi.org/10.48550/arxiv.2309.10158" @default.
- W4386907174 hasPublicationYear "2023" @default.
- W4386907174 type Work @default.
- W4386907174 citedByCount "0" @default.
- W4386907174 crossrefType "posted-content" @default.
- W4386907174 hasAuthorship W4386907174A5006952036 @default.
- W4386907174 hasAuthorship W4386907174A5080429518 @default.
- W4386907174 hasBestOaLocation W43869071741 @default.
- W4386907174 hasConcept C112640561 @default.
- W4386907174 hasConcept C153180895 @default.
- W4386907174 hasConcept C154945302 @default.
- W4386907174 hasConcept C204321447 @default.
- W4386907174 hasConcept C2779386606 @default.
- W4386907174 hasConcept C28490314 @default.
- W4386907174 hasConcept C41008148 @default.
- W4386907174 hasConcept C52622490 @default.
- W4386907174 hasConcept C95623464 @default.
- W4386907174 hasConceptScore W4386907174C112640561 @default.
- W4386907174 hasConceptScore W4386907174C153180895 @default.
- W4386907174 hasConceptScore W4386907174C154945302 @default.
- W4386907174 hasConceptScore W4386907174C204321447 @default.
- W4386907174 hasConceptScore W4386907174C2779386606 @default.
- W4386907174 hasConceptScore W4386907174C28490314 @default.
- W4386907174 hasConceptScore W4386907174C41008148 @default.
- W4386907174 hasConceptScore W4386907174C52622490 @default.
- W4386907174 hasConceptScore W4386907174C95623464 @default.
- W4386907174 hasLocation W43869071741 @default.
- W4386907174 hasOpenAccess W4386907174 @default.
- W4386907174 hasPrimaryLocation W43869071741 @default.
- W4386907174 hasRelatedWork W2136489243 @default.
- W4386907174 hasRelatedWork W2137744223 @default.
- W4386907174 hasRelatedWork W2144059113 @default.
- W4386907174 hasRelatedWork W2146076056 @default.
- W4386907174 hasRelatedWork W2158785961 @default.
- W4386907174 hasRelatedWork W2563096758 @default.
- W4386907174 hasRelatedWork W2811390910 @default.
- W4386907174 hasRelatedWork W2888573161 @default.
- W4386907174 hasRelatedWork W3003836766 @default.
- W4386907174 hasRelatedWork W3122954453 @default.
- W4386907174 isParatext "false" @default.
- W4386907174 isRetracted "false" @default.
- W4386907174 workType "article" @default.