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 2021-03-04 13:30
[Poster Presentation] Object Boundary Correction for Monocular Depth Estimation Using Region Segmentation
Ikuma Yasukawa, Shoko Imaizumi (Chiba Univ.) EMM2020-67
Abstract (in Japanese) (See Japanese page) 
(in English) We propose a new method to generate high-quality depth images in this paper. The proposed method corrects object boundaries in a depth image derived from the monocular image using deep learning.A large amount of depth data is required for depth estimation with deep learning. However, it is difficult to fully prepare the depth data with the same size as the original monocular image. Therefore, a depth estimation method by self-supervised learning, which introduces stereo video sequences as training data, has been proposed.This method utilizes disparities of stereo video for learning, and thus the object boundaries in the estimated depth image tend to be ambiguous. In contrast, the proposed method first generates a background depth image without target objects and an object depth image of the target objects separately, and then combines both depth images. We evaluate the quality of the depth images obtained by the proposed and conventional methods and confirm the effectiveness of our method.
Keyword (in Japanese) (See Japanese page) 
(in English) Depth estimation / boundary correction / segmentation / inpainting / multi-view image generation / / /  
Reference Info. IEICE Tech. Rep., vol. 120, no. 418, EMM2020-67, pp. 1-6, March 2021.
Paper # EMM2020-67 
Date of Issue 2021-02-25 (EMM) 
ISSN Print edition: ISSN 0913-5685  Online edition: ISSN 2432-6380
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 EMM2020-67

Conference Information
Committee EMM  
Conference Date 2021-03-04 - 2021-03-05 
Place (in Japanese) (See Japanese page) 
Place (in English) Online 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Image and Sound Quality, Metrics for Perception and Recognition, Human Auditory and Visual System, etc. 
Paper Information
Registration To EMM 
Conference Code 2021-03-EMM 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Object Boundary Correction for Monocular Depth Estimation Using Region Segmentation 
Sub Title (in English)  
Keyword(1) Depth estimation  
Keyword(2) boundary correction  
Keyword(3) segmentation  
Keyword(4) inpainting  
Keyword(5) multi-view image generation  
1st Author's Name Ikuma Yasukawa  
1st Author's Affiliation Chiba University (Chiba Univ.)
2nd Author's Name Shoko Imaizumi  
2nd Author's Affiliation Chiba University (Chiba 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 ()
Date Time 2021-03-04 13:30:00 
Presentation Time 15 
Registration for EMM 
Paper # IEICE-EMM2020-67 
Volume (vol) IEICE-120 
Number (no) no.418 
Page pp.1-6 
#Pages IEICE-6 
Date of Issue IEICE-EMM-2021-02-25 

[Return to Top Page]

[Return to IEICE Web Page]

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