IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
Tech. Rep. Archives
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2023-03-02 17:00
A Semi-Supervised Learning Framework for Handwritten Text Recognition using Mixed Augmentations and Scheduled Pseudo-Label Loss
Masayuki Honda, Hung Tuan Nguyen, Cuong Tuan Nguyen (TUAT), Cong Kha Nguyen, Ryosuke Odate, Takashi Kanemaru (Hitachi Ltd.), Masaki Nakagawa (TUAT) PRMU2022-97 IBISML2022-104
Abstract (in Japanese) (See Japanese page) 
(in English) We propose Incremental Teacher Model, a semi-supervised learning (SSL) framework for handwriting text recognition. The framework comprises a teacher model and a student model learning through Scheduled Pseudo-Label loss with Mixed Augmentations. First, the student model is pre-trained by labeled samples and used to initiate the teacher model. The student model is further trained on transformed samples provided by Mixed Augmentations. This training process uses pseudo-labels generated by the teacher model. After a training epoch, the teacher model is updated from a well-validated student model. We apply the proposed framework to four architectures of different handwriting recognizers. For almost every architecture, the recognizer trained by Incremental Teacher Model outperforms the recognizers trained by the well-known SSL methods on the IAM handwriting database.
Keyword (in Japanese) (See Japanese page) 
(in English) Semi-Supervised Learning / Training framework / Handwriting recognition / / / / /  
Reference Info. IEICE Tech. Rep., vol. 122, no. 404, PRMU2022-97, pp. 199-204, March 2023.
Paper # PRMU2022-97 
Date of Issue 2023-02-23 (PRMU, IBISML) 
ISSN Online edition: ISSN 2432-6380
Copyright
and
reproduction
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034)
Download PDF PRMU2022-97 IBISML2022-104

Conference Information
Committee PRMU IBISML IPSJ-CVIM  
Conference Date 2023-03-02 - 2023-03-03 
Place (in Japanese) (See Japanese page) 
Place (in English) Future University Hakodate 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To PRMU 
Conference Code 2023-03-PRMU-IBISML-CVIM 
Language English 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) A Semi-Supervised Learning Framework for Handwritten Text Recognition using Mixed Augmentations and Scheduled Pseudo-Label Loss 
Sub Title (in English)  
Keyword(1) Semi-Supervised Learning  
Keyword(2) Training framework  
Keyword(3) Handwriting recognition  
Keyword(4)  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Masayuki Honda  
1st Author's Affiliation Tokyo University of Agriculture and Technology (TUAT)
2nd Author's Name Hung Tuan Nguyen  
2nd Author's Affiliation Tokyo University of Agriculture and Technology (TUAT)
3rd Author's Name Cuong Tuan Nguyen  
3rd Author's Affiliation Tokyo University of Agriculture and Technology (TUAT)
4th Author's Name Cong Kha Nguyen  
4th Author's Affiliation Hitachi Ltd. (Hitachi Ltd.)
5th Author's Name Ryosuke Odate  
5th Author's Affiliation Hitachi Ltd. (Hitachi Ltd.)
6th Author's Name Takashi Kanemaru  
6th Author's Affiliation Hitachi Ltd. (Hitachi Ltd.)
7th Author's Name Masaki Nakagawa  
7th Author's Affiliation Tokyo University of Agriculture and Technology (TUAT)
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
Speaker Author-1 
Date Time 2023-03-02 17:00:00 
Presentation Time 10 minutes 
Registration for PRMU 
Paper # PRMU2022-97, IBISML2022-104 
Volume (vol) vol.122 
Number (no) no.404(PRMU), no.405(IBISML) 
Page pp.199-204 
#Pages
Date of Issue 2023-02-23 (PRMU, IBISML) 


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan