Openly accessible

An end-to-end text spotter with text relation networks

Jiang, J, Wei, B, Yu, M, Li, Gang, Li, B, Liu, C, Li, M and Huang, W 2021, An end-to-end text spotter with text relation networks, Cybersecurity, vol. 4, pp. 1-13, doi: 10.1186/s42400-021-00073-x.

Attached Files
Name Description MIMEType Size Downloads

Title An end-to-end text spotter with text relation networks
Author(s) Jiang, J
Wei, B
Yu, M
Li, GangORCID iD for Li, Gang orcid.org/0000-0003-1583-641X
Li, B
Liu, C
Li, M
Huang, W
Journal name Cybersecurity
Volume number 4
Article ID 7
Start page 1
End page 13
Total pages 13
Publisher Springer
Place of publication Berlin, Germany
Publication date 2021
ISSN 2096-4862
2523-3246
Keyword(s) Scene text spotting
Graph convolutional network
Visual reasoning
Summary Reading text in images automatically has become an attractive research topic in computer vision. Specifically, end-to-end spotting of scene text has attracted significant research attention, and relatively ideal accuracy has been achieved on several datasets. However, most of the existing works overlooked the semantic connection between the scene text instances, and had limitations in situations such as occlusion, blurring, and unseen characters, which result in some semantic information lost in the text regions. The relevance between texts generally lies in the scene images. From the perspective of cognitive psychology, humans often combine the nearby easy-to-recognize texts to infer the unidentifiable text. In this paper, we propose a novel graph-based method for intermediate semantic features enhancement, called Text Relation Networks. Specifically, we model the co-occurrence relationship of scene texts as a graph. The nodes in the graph represent the text instances in a scene image, and the corresponding semantic features are defined as representations of the nodes. The relative positions between text instances are measured as the weights of edges in the established graph. Then, a convolution operation is performed on the graph to aggregate semantic information and enhance the intermediate features corresponding to text instances. We evaluate the proposed method through comprehensive experiments on several mainstream benchmarks, and get highly competitive results. For example, on the , our method surpasses the previous top works by 2.1% on the word spotting task.
Language eng
DOI 10.1186/s42400-021-00073-x
Indigenous content off
HERDC Research category C1 Refereed article in a scholarly journal
Free to Read? Yes
Persistent URL http://hdl.handle.net/10536/DRO/DU:30150446

Connect to link resolver
 
Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved.

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.

Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in TR Web of Science
Scopus Citation Count Cited 0 times in Scopus
Google Scholar Search Google Scholar
Access Statistics: 7 Abstract Views, 0 File Downloads  -  Detailed Statistics
Created: Mon, 03 May 2021, 13:19:53 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.