Reading Text in the Wild from Compressed Images

Leonardo Galteri, Dena Bazazian, Lorenzo Seidenari, Marco Bertini, Andrew D. Bagdanov, Anguelos Nicolaou, Dimosthenis Karatzas, Alberto Del Bimbo

Producció científica: Capítol de llibreCapítolRecercaAvaluat per experts

2 Cites (Scopus)

Resum

Reading text in the wild is gaining attention in the computer vision community. Images captured in the wild are almost always compressed to varying degrees, depending on application context, and this compression introduces artifacts that distort image content into the captured images. In this paper we investigate the impact these compression artifacts have on text localization and recognition in the wild. We also propose a deep Convolutional Neural Network (CNN) that can eliminate text-specific compression artifacts and which leads to an improvement in text recognition. Experimental results on the ICDAR-Challenge4 dataset demonstrate that compression artifacts have a significant impact on text localization and recognition and that our approach yields an improvement in both - especially at high compression rates.

Idioma originalAnglès
Títol de la publicacióProceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017
EditorInstitute of Electrical and Electronics Engineers Inc.
Pàgines2399-2407
Nombre de pàgines9
ISBN (electrònic)9781538610343
DOIs
Estat de la publicacióPublicada - 1 de jul. 2017

Sèrie de publicacions

NomProceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017
Volum2018-January

Fingerprint

Navegar pels temes de recerca de 'Reading Text in the Wild from Compressed Images'. Junts formen un fingerprint únic.

Com citar-ho