DeepText: Detecting text from the wild with multi-ASPP-assembled deeplab

Wang, Q; Jia, W; He, X; Lu, Y; Blumenstein, M; Huang, Y; Lyu, S

DeepText: Detecting text from the wild with multi-ASPP-assembled deeplab

Wang, Q Jia, W

He, X

Lu, Y Blumenstein, M

Huang, Y Lyu, S

Permalink

Publisher:: IEEE
Publication Type:: Conference Proceeding
Citation:: Proceedings of the International Conference on Document Analysis and Recognition, ICDAR, 2020, 00, pp. 208-213
Issue Date:: 2020

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

The embargo period expires on 3 Feb 2022

Adobe PDF

Download Accepted versionAdobe PDF (729.56 kB)

View on publisher's site

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Wang, Q
dc.contributor.author	Jia, W https://orcid.org/0000-0002-0940-3338
dc.contributor.author	He, X https://orcid.org/0000-0001-8962-540X
dc.contributor.author	Lu, Y
dc.contributor.author	Blumenstein, M https://orcid.org/0000-0002-9908-3744
dc.contributor.author	Huang, Y
dc.contributor.author	Lyu, S
dc.date	2019-09-20
dc.date.accessioned	2020-11-03T23:12:27Z
dc.date.available	2020-11-03T23:12:27Z
dc.date.issued	2020
dc.identifier.citation	Proceedings of the International Conference on Document Analysis and Recognition, ICDAR, 2020, 00, pp. 208-213
dc.identifier.isbn	9781728128610
dc.identifier.issn	1520-5363
dc.identifier.uri	http://hdl.handle.net/10453/143739
dc.description.abstract	© 2019 IEEE. In this paper, we address the issue of scene text detection in the way of direct regression and successfully adapt an effective semantic segmentation model, DeepLab v3+ [1], for this application. In order to handle texts with arbitrary orientations and sizes and improve the recall of small texts, we propose to extract features of multiple scales by inserting multiple Atrous Spatial Pyramid Pooling (ASPP) layers to the DeepLab after the feature maps with different resolutions. Then, we set multiple auxiliary IoU losses at the decoding stage and make auxiliary connections from the intermediate encoding layers to the decoder to assist network training and enhance the discrimination ability of lower encoding layers. Experiments conducted on the benchmark scene text dataset ICDAR2015 demonstrate the superior performance of our proposed network, named as DeepText, over the state-of-the-art approaches.
dc.language	en
dc.publisher	IEEE
dc.relation.ispartof	Proceedings of the International Conference on Document Analysis and Recognition, ICDAR
dc.relation.ispartof	International Conference on Document Analysis and Recognition
dc.relation.hasversion	Accepted version
dc.relation.isbasedon	10.1109/ICDAR.2019.00042
dc.rights	info:eu-repo/semantics/embargoedAccess
dc.rights	© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.	en_US
dc.title	DeepText: Detecting text from the wild with multi-ASPP-assembled deeplab
dc.type	Conference Proceeding
utslib.citation.volume	00
utslib.location.activity	Sydney, Australia
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Strength - CRIN - Realtime Information Networks
pubs.organisational-group	/University of Technology Sydney/Strength - GBDTC - Global Big Data Technologies
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Electrical and Data Engineering
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Strength - AAII - Australian Artificial Intelligence Institute
pubs.organisational-group	/University of Technology Sydney/Strength - QSI - Centre for Quantum Software and Information
utslib.copyright.status	open_access	*
pubs.consider-herdc	false
utslib.copyright.embargo	2022-02-03T00:00:00+1000Z
dc.date.updated	2020-11-03T23:12:23Z
pubs.finish-date	2019-09-25
pubs.place-of-publication	Piscataway, USA
pubs.publication-status	Published
pubs.start-date	2019-09-20
pubs.volume	00
dc.location	Piscataway, USA

Abstract:

© 2019 IEEE. In this paper, we address the issue of scene text detection in the way of direct regression and successfully adapt an effective semantic segmentation model, DeepLab v3+ [1], for this application. In order to handle texts with arbitrary orientations and sizes and improve the recall of small texts, we propose to extract features of multiple scales by inserting multiple Atrous Spatial Pyramid Pooling (ASPP) layers to the DeepLab after the feature maps with different resolutions. Then, we set multiple auxiliary IoU losses at the decoding stage and make auxiliary connections from the intermediate encoding layers to the decoder to assist network training and enhance the discrimination ability of lower encoding layers. Experiments conducted on the benchmark scene text dataset ICDAR2015 demonstrate the superior performance of our proposed network, named as DeepText, over the state-of-the-art approaches.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/143739