Publication Details

TS-Net: OCR Trained to Switch Between Text Transcription Styles

KOHÚT, J.; HRADIŠ, M. TS-Net: OCR Trained to Switch Between Text Transcription Styles. In Lladós J., Lopresti D., Uchida S. (eds) Document Analysis and Recognition - ICDAR 2021. Lecture Notes in Computer Science. Lecture Notes in Computer Science. Lausanne: Springer Nature Switzerland AG, 2021. p. 478-493. ISBN: 978-3-030-86336-4. ISSN: 0302-9743.
Czech title
TS-Net: OCR pro přepínání stylu přepisu
Type
conference paper
Language
English
Authors
Kohút Jan, Ing. (DCGM)
Hradiš Michal, Ing., Ph.D. (DCGM)
URL
Keywords

Transcription styles, Adaptive instance normalization, Text recognition, Neural
networks, CTC

Abstract

Multiple transcribers produce transcriptions in inconsistent transcription
styles. 
This presents a problem for training consistent neural network systems for text
recognition.
We propose Transcription Style Block (TSB) which can learn to switch between
multiple transcription styles without any explicit knowledge about the
transcription rules.
TSB is an adaptive instance normalization conditioned by transcription style
identifiers e.g. document numbers or transcriber names and it can be added near
the end of any standard text recognition network. 
We show that TSB is robust towards the number and complexity of transcription
styles and does not degrade the text recognition performance.
With time and data efficient adaptation to a new transcription style, we achieved
up to 77\% relative test character error reduction in comparison to a network
without the TSB. 

Published
2021
Pages
478–493
Journal
Lecture Notes in Computer Science, vol. 12824, no. 1, ISSN 0302-9743
Proceedings
Lladós J., Lopresti D., Uchida S. (eds) Document Analysis and Recognition - ICDAR 2021
Series
Lecture Notes in Computer Science
Conference
International Conference on Document Analysis and Recognition, Lausanne, Switzerland, CH
ISBN
978-3-030-86336-4
Publisher
Springer Nature Switzerland AG
Place
Lausanne
DOI
10.1007/978-3-030-86337-1_32
UT WoS
000711880100032
EID Scopus
2-s2.0-85115345936
BibTeX 
@inproceedings{BUT169806,
  author="Jan {Kohút} and Michal {Hradiš}",
  title="TS-Net: OCR Trained to Switch Between Text Transcription Styles",
  booktitle="Lladós J., Lopresti D., Uchida S. (eds) Document Analysis and Recognition - ICDAR 2021",
  year="2021",
  series="Lecture Notes in Computer Science",
  journal="Lecture Notes in Computer Science",
  volume="12824",
  number="1",
  pages="478--493",
  publisher="Springer Nature Switzerland AG",
  address="Lausanne",
  doi="10.1007/978-3-030-86337-1\{_}32",
  isbn="978-3-030-86336-4",
  issn="0302-9743",
  url="https://pero.fit.vutbr.cz/publications"
}
Back to top