CASIA OpenIR  > 模式识别国家重点实验室  > 模式分析与学习
Handwritten Text Recognition with Convolutional Prototype Network and Most Aligned Frame Based CTC Training
Likun Gao1,2; Heng Zhang1; Cheng-Lin Liu1,2,3
2021-09
Conference Name16th International Conference on Document Analysis and Recognition
Conference Date2021-9-5
Conference PlaceLausanne, Switzerland
PublisherSpringer
Abstract

End-to-end Frameworks with Connectionist Temporal Clas-
sification (CTC) have achieved great success in text recognition. Despite
high accuracies with deep learning, CTC-based text recognition meth-
ods also suffer from poor alignment (character boundary positioning)
in many applications. To address this issue, we propose an end-to-end
text recognition method based on robust prototype learning. In the new
CTC framework, we formulate the blank as the rejection of character
classes and use the one-vs-all prototype classifier as the output layer of
the convolutional neural network. For network learning, based on forced
alignment between frames and character labels, the most aligned frame
is up-weighted in CTC training strategy to reduce estimation errors in
decoding. Experiments of handwritten text recognition on four bench-
mark datasets of different languages show that the proposed method
consistently improves the accuracy and alignment of CTC-based text
recognition baseline.

Indexed ByEI
Language英语
Document Type会议论文
Identifierhttp://ir.ia.ac.cn/handle/173211/45027
Collection模式识别国家重点实验室_模式分析与学习
Corresponding AuthorCheng-Lin Liu
Affiliation1.National Laboratory of Pattern Recognition (NLPR), Institution of Automation, Chinese Academy of Sciences, Beijing 100190, China
2.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China
3.CAS Center for Excellence in Brain Science and Intelligence Technology, Beijing 100190, China
First Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Corresponding Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Recommended Citation
GB/T 7714
Likun Gao,Heng Zhang,Cheng-Lin Liu. Handwritten Text Recognition with Convolutional Prototype Network and Most Aligned Frame Based CTC Training[C]:Springer,2021.
Files in This Item: Download All
File Name/Size DocType Version Access License
main.pdf(1804KB)会议论文 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Likun Gao]'s Articles
[Heng Zhang]'s Articles
[Cheng-Lin Liu]'s Articles
Baidu academic
Similar articles in Baidu academic
[Likun Gao]'s Articles
[Heng Zhang]'s Articles
[Cheng-Lin Liu]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Likun Gao]'s Articles
[Heng Zhang]'s Articles
[Cheng-Lin Liu]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: main.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.