How salient is scene text?

Asif Shahab, Faisal Shafait, Andreas Dengel, Seiichi Uchida

Research output: Chapter in Book/Report/Conference proceedingConference contribution

25 Citations (Scopus)

Abstract

Computational models of visual attention use image features to identify salient locations in an image that are likely to attract human attention. Attention models have been quite effectively used for various object detection tasks. However, their use for scene text detection is under-investigated. As a general observation, scene text often conveys important information and is usually prominent or salient in the scene itself. In this paper, we evaluate four state-of-the-art attention models for their response to scene text. Initial results indicate that saliency maps produced by these attention models can be used for aiding scene text detection algorithms by suppressing non-text regions.

Original languageEnglish
Title of host publicationProceedings - 10th IAPR International Workshop on Document Analysis Systems, DAS 2012
Pages317-321
Number of pages5
DOIs
Publication statusPublished - 2012
Event10th IAPR International Workshop on Document Analysis Systems, DAS 2012 - Gold Coast, QLD, Australia
Duration: Mar 27 2012Mar 29 2012

Other

Other10th IAPR International Workshop on Document Analysis Systems, DAS 2012
CountryAustralia
CityGold Coast, QLD
Period3/27/123/29/12

Fingerprint

Object detection

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering

Cite this

Shahab, A., Shafait, F., Dengel, A., & Uchida, S. (2012). How salient is scene text? In Proceedings - 10th IAPR International Workshop on Document Analysis Systems, DAS 2012 (pp. 317-321). [6195386] https://doi.org/10.1109/DAS.2012.42

How salient is scene text? / Shahab, Asif; Shafait, Faisal; Dengel, Andreas; Uchida, Seiichi.

Proceedings - 10th IAPR International Workshop on Document Analysis Systems, DAS 2012. 2012. p. 317-321 6195386.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Shahab, A, Shafait, F, Dengel, A & Uchida, S 2012, How salient is scene text? in Proceedings - 10th IAPR International Workshop on Document Analysis Systems, DAS 2012., 6195386, pp. 317-321, 10th IAPR International Workshop on Document Analysis Systems, DAS 2012, Gold Coast, QLD, Australia, 3/27/12. https://doi.org/10.1109/DAS.2012.42
Shahab A, Shafait F, Dengel A, Uchida S. How salient is scene text? In Proceedings - 10th IAPR International Workshop on Document Analysis Systems, DAS 2012. 2012. p. 317-321. 6195386 https://doi.org/10.1109/DAS.2012.42
Shahab, Asif ; Shafait, Faisal ; Dengel, Andreas ; Uchida, Seiichi. / How salient is scene text?. Proceedings - 10th IAPR International Workshop on Document Analysis Systems, DAS 2012. 2012. pp. 317-321
@inproceedings{2474c0fe5629470d8794dd215596dfb3,
title = "How salient is scene text?",
abstract = "Computational models of visual attention use image features to identify salient locations in an image that are likely to attract human attention. Attention models have been quite effectively used for various object detection tasks. However, their use for scene text detection is under-investigated. As a general observation, scene text often conveys important information and is usually prominent or salient in the scene itself. In this paper, we evaluate four state-of-the-art attention models for their response to scene text. Initial results indicate that saliency maps produced by these attention models can be used for aiding scene text detection algorithms by suppressing non-text regions.",
author = "Asif Shahab and Faisal Shafait and Andreas Dengel and Seiichi Uchida",
year = "2012",
doi = "10.1109/DAS.2012.42",
language = "English",
isbn = "9780769546612",
pages = "317--321",
booktitle = "Proceedings - 10th IAPR International Workshop on Document Analysis Systems, DAS 2012",

}

TY - GEN

T1 - How salient is scene text?

AU - Shahab, Asif

AU - Shafait, Faisal

AU - Dengel, Andreas

AU - Uchida, Seiichi

PY - 2012

Y1 - 2012

N2 - Computational models of visual attention use image features to identify salient locations in an image that are likely to attract human attention. Attention models have been quite effectively used for various object detection tasks. However, their use for scene text detection is under-investigated. As a general observation, scene text often conveys important information and is usually prominent or salient in the scene itself. In this paper, we evaluate four state-of-the-art attention models for their response to scene text. Initial results indicate that saliency maps produced by these attention models can be used for aiding scene text detection algorithms by suppressing non-text regions.

AB - Computational models of visual attention use image features to identify salient locations in an image that are likely to attract human attention. Attention models have been quite effectively used for various object detection tasks. However, their use for scene text detection is under-investigated. As a general observation, scene text often conveys important information and is usually prominent or salient in the scene itself. In this paper, we evaluate four state-of-the-art attention models for their response to scene text. Initial results indicate that saliency maps produced by these attention models can be used for aiding scene text detection algorithms by suppressing non-text regions.

UR - http://www.scopus.com/inward/record.url?scp=84862101874&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862101874&partnerID=8YFLogxK

U2 - 10.1109/DAS.2012.42

DO - 10.1109/DAS.2012.42

M3 - Conference contribution

AN - SCOPUS:84862101874

SN - 9780769546612

SP - 317

EP - 321

BT - Proceedings - 10th IAPR International Workshop on Document Analysis Systems, DAS 2012

ER -