Information: Join today and make your research activities more affordable! Technical workshop participation fees and annual registration fees are available at member rates.
Notice: [Important] Announcement of Changes to Registration Fee Payment and Manuscript Upload Procedures for IEICE Technical Meetings
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 2022-12-16 10:15
Pose-aware Disentangled Multiscale Transformer for Pose Guided Person Image Generation
Kei Shibasaki, Masaaki Ikehara (Keio Univ.) PRMU2022-44
Abstract (in Japanese) (See Japanese page) 
(in English) Pose Guided Person Image Generation (PGPIG) is the task that transforms the pose of a person image from the source image, its pose information and the target pose information. Most existing PGPIG methods require additional pose information or tasks, limiting their application. In addition, all input information is combined and fed into the network, and CNNs are used as the feature extractor. However, CNNs can only extract features from neighboring pixels and cannot consider the consistency of the entire image. Furthermore, they combine the input information before extracting enough features, making it unclear which task the network should learn, which degrades the network performance. This paper proposes a PGPIG network that addresses the image consistency problem and clarifies which task the network should learn. The proposed method disentangles the PGPIG task into two sub tasks: “rough pose transformation” and “detailed texture generation”. In the former task, low-resolution feature maps are transformed by blocks containing Axial Transformer with a large receptive field. These blocks employ an Encoder-Decoder structure, which allows the network to use the pose information well and improves the stability and performance of the training. The latter task uses a CNN network with Adaptive Instance Normalization. Experiments show that the proposed method has competitive performance with other state-of-the-art methods. Furthermore, despite achieving excellent performance, the proposed network has a significantly fewer parameters than existing methods.
Keyword (in Japanese) (See Japanese page) 
(in English) Deep learning / Image Processing / Pose Guided Person Image Generation / Transformer / Multi-scale Network / / /  
Reference Info. IEICE Tech. Rep., vol. 122, no. 314, PRMU2022-44, pp. 63-69, Dec. 2022.
Paper # PRMU2022-44 
Date of Issue 2022-12-08 (PRMU) 
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 PRMU2022-44

Conference Information
Committee PRMU  
Conference Date 2022-12-15 - 2022-12-16 
Place (in Japanese) (See Japanese page) 
Place (in English) Toyama International Conference Center 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To PRMU 
Conference Code 2022-12-PRMU 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Pose-aware Disentangled Multiscale Transformer for Pose Guided Person Image Generation 
Sub Title (in English)  
Keyword(1) Deep learning  
Keyword(2) Image Processing  
Keyword(3) Pose Guided Person Image Generation  
Keyword(4) Transformer  
Keyword(5) Multi-scale Network  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Kei Shibasaki  
1st Author's Affiliation Keio University (Keio Univ.)
2nd Author's Name Masaaki Ikehara  
2nd Author's Affiliation Keio University (Keio 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 ()
21st Author's Name  
21st Author's Affiliation ()
22nd Author's Name  
22nd Author's Affiliation ()
23rd Author's Name  
23rd Author's Affiliation ()
24th Author's Name  
24th Author's Affiliation ()
25th Author's Name  
25th Author's Affiliation ()
26th Author's Name / /
26th Author's Affiliation ()
()
27th Author's Name / /
27th Author's Affiliation ()
()
28th Author's Name / /
28th Author's Affiliation ()
()
29th Author's Name / /
29th Author's Affiliation ()
()
30th Author's Name / /
30th Author's Affiliation ()
()
31st Author's Name / /
31st Author's Affiliation ()
()
32nd Author's Name / /
32nd Author's Affiliation ()
()
33rd Author's Name / /
33rd Author's Affiliation ()
()
34th Author's Name / /
34th Author's Affiliation ()
()
35th Author's Name / /
35th Author's Affiliation ()
()
36th Author's Name / /
36th Author's Affiliation ()
()
Speaker Author-1 
Date Time 2022-12-16 10:15:00 
Presentation Time 15 minutes 
Registration for PRMU 
Paper # PRMU2022-44 
Volume (vol) vol.122 
Number (no) no.314 
Page pp.63-69 
#Pages 7 
Date of Issue 2022-12-08 (PRMU) 

[Return to Top Page]

[Return to IEICE Web Page]
The Institute of Electronics, Information and Communication Engineers (IEICE), Japan