ITE Technical Group Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2021-06-04 15:15
Acceleration of Dictionary Learning by Non-Separable Oversampled Lapped Transform for Image Restoration
Soushi Takahashi, Shogo Muramatsu (Niigata Univ.)
Abstract (in Japanese) (See Japanese page) 
(in English) This study proposes to accelerate example-based design of non-separable oversampled lapped transform (NSOLT) for image restoration. With development of the recent measurement techniques, it becomes possible to obtain a various huge amount of data. At the same time, image restoration by sparse modeling is demanded. There is a dictionary learning technique as an example of image restoration techniques. Muramatsu et al. proposed convolutional dictionary learning with NSOLT and showed the significance. Stochastic gradient decent (SGD) is used for optimization of design parameters on NSOLT as a structured convolutional dictionary. However, SGD has a disadvantage at convergence, so there is room for improvement. In this report, we propose to apply stochastic variance reduced gradient (SVRG) which is a variant of SGD and review the acceleration performance of the dictionary learning.
Keyword (in Japanese) (See Japanese page) 
(in English) Dictionary learning / Sparse modeling / Non-separable oversampled lapped transform / Stochastic variance reduced gradient / Image restoration / / /  
Reference Info. ITE Tech. Rep., vol. 45, no. 13, ME2021-58, pp. 87-92, June 2021.
Paper # ME2021-58 
Date of Issue 2021-05-27 (IST, ME) 
ISSN Print edition: ISSN 1342-6893  Online edition: ISSN 2424-1970
Download PDF

Conference Information
Committee IST ME IEICE-IE IEICE-SIP IEICE-BioX  
Conference Date 2021-06-03 - 2021-06-04 
Place (in Japanese) (See Japanese page) 
Place (in English) Online 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To ME 
Conference Code 2021-06-IST-ME-IE-SIP-BioX 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Acceleration of Dictionary Learning by Non-Separable Oversampled Lapped Transform for Image Restoration 
Sub Title (in English)  
Keyword(1) Dictionary learning  
Keyword(2) Sparse modeling  
Keyword(3) Non-separable oversampled lapped transform  
Keyword(4) Stochastic variance reduced gradient  
Keyword(5) Image restoration  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Soushi Takahashi  
1st Author's Affiliation Niigata University (Niigata Univ.)
2nd Author's Name Shogo Muramatsu  
2nd Author's Affiliation Niigata University (Niigata Univ.)
3rd Author's Name  
3rd Author's Affiliation ()
4th Author's Name  
4th Author's Affiliation ()
5th Author's Name  
5th Author's Affiliation ()
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
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
Date Time 2021-06-04 15:15:00 
Presentation Time 30 
Registration for ME 
Paper # ITE-IST2021-27,ITE-ME2021-58 
Volume (vol) ITE-45 
Number (no) no.13 
Page pp.87-92 
#Pages ITE-6 
Date of Issue ITE-IST-2021-05-27,ITE-ME-2021-05-27 


[Return to Top Page]

[Return to ITE Web Page]


The Institute of Image Information and Television Engineers (ITE), Japan