AT-ST: Self-Training Adaptation Strategy for OCR in Domains with Limited Transcriptions

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F21%3APU142899" target="_blank" >RIV/00216305:26230/21:PU142899 - isvavai.cz</a>

Výsledek na webu

<a href="https://pero.fit.vutbr.cz/publications" target="_blank" >https://pero.fit.vutbr.cz/publications</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-030-86337-1_31" target="_blank" >10.1007/978-3-030-86337-1_31</a>

Jazyk výsledku

angličtina

Název v původním jazyce

AT-ST: Self-Training Adaptation Strategy for OCR in Domains with Limited Transcriptions

Popis výsledku v původním jazyce

This paper addresses text recognition for domains with limited manual annotations by a simple self-training strategy. Our approach should reduce human annotation effort when target domain data is plentiful, such as when transcribing a collection of single person's correspondence or a large manuscript. We propose to train a seed system on large scale data from related domains mixed with available annotated data from the target domain. The seed system transcribes the unannotated data from the target domain which is then used to train a better system. We study several confidence measures and eventually decide to use the posterior probability of a transcription for data selection. Additionally, we propose to augment the data using an aggressive masking scheme. By self-training, we achieve up to 55 % reduction in character error rate for handwritten data and up to 38 % on printed data. The masking augmentation itself reduces the error rate by about 10 % and its effect is better pronounced in case of difficult handwritten data.

Název v anglickém jazyce

AT-ST: Self-Training Adaptation Strategy for OCR in Domains with Limited Transcriptions

Popis výsledku anglicky

This paper addresses text recognition for domains with limited manual annotations by a simple self-training strategy. Our approach should reduce human annotation effort when target domain data is plentiful, such as when transcribing a collection of single person's correspondence or a large manuscript. We propose to train a seed system on large scale data from related domains mixed with available annotated data from the target domain. The seed system transcribes the unannotated data from the target domain which is then used to train a better system. We study several confidence measures and eventually decide to use the posterior probability of a transcription for data selection. Additionally, we propose to augment the data using an aggressive masking scheme. By self-training, we achieve up to 55 % reduction in character error rate for handwritten data and up to 38 % on printed data. The masking augmentation itself reduces the error rate by about 10 % and its effect is better pronounced in case of difficult handwritten data.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2021

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Lladós J., Lopresti D., Uchida S. (eds) Document Analysis and Recognition - ICDAR 2021

ISBN

978-3-030-86336-4

ISSN

—

e-ISSN

—

Počet stran výsledku

14

Strana od-do

463-477

Název nakladatele

Springer Nature Switzerland AG

Místo vydání

Lausanne

Místo konání akce

Lausanne, Switzerland

Datum konání akce

5. 9. 2021

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000711880100031