Scale Invariant Multi-view Depth Estimation Network with cGAN Refinement

Chia Hung Yeh; Yao Pao Huang; Mei Juan Chen

doi:10.1007/978-981-13-9190-3_75

Scale Invariant Multi-view Depth Estimation Network with cGAN Refinement

Chia Hung Yeh^*, Yao Pao Huang, Mei Juan Chen

^*Corresponding author for this work

Department of Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper we propose a deep learning based depth estimation method for monocular RGB sequences. We train a pair of encoder-decoder network to resolve depth information form image pairs and relative camera poses. To solve scale ambiguous of monocular sequences, a conditional generative adversarial network is applied. Experimental results show that the proposed method can overcome the problem of scale ambiguous and therefore is more suitable for a variety of applications.

Original language	English
Title of host publication	New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers
Editors	Chuan-Yu Chang, Chien-Chou Lin, Horng-Horng Lin
Publisher	Springer Verlag
Pages	681-687
Number of pages	7
ISBN (Print)	9789811391897
DOIs	https://doi.org/10.1007/978-981-13-9190-3_75
Publication status	Published - 2019
Event	23rd International Computer Symposium, ICS 2018 - Yunlin, Taiwan Duration: 2018 Dec 20 → 2018 Dec 22

Publication series

Name	Communications in Computer and Information Science
Volume	1013
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	23rd International Computer Symposium, ICS 2018
Country/Territory	Taiwan
City	Yunlin
Period	2018/12/20 → 2018/12/22

Keywords

Conditional generative adversarial network
Deep learning
Multi-view depth estimation

ASJC Scopus subject areas

General Computer Science
General Mathematics

Access to Document

10.1007/978-981-13-9190-3_75

Cite this

Yeh, C. H., Huang, Y. P., & Chen, M. J. (2019). Scale Invariant Multi-view Depth Estimation Network with cGAN Refinement. In C.-Y. Chang, C.-C. Lin, & H.-H. Lin (Eds.), New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers (pp. 681-687). (Communications in Computer and Information Science; Vol. 1013). Springer Verlag. https://doi.org/10.1007/978-981-13-9190-3_75

Scale Invariant Multi-view Depth Estimation Network with cGAN Refinement. / Yeh, Chia Hung; Huang, Yao Pao; Chen, Mei Juan.
New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers. ed. / Chuan-Yu Chang; Chien-Chou Lin; Horng-Horng Lin. Springer Verlag, 2019. p. 681-687 (Communications in Computer and Information Science; Vol. 1013).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Yeh, CH, Huang, YP & Chen, MJ 2019, Scale Invariant Multi-view Depth Estimation Network with cGAN Refinement. in C-Y Chang, C-C Lin & H-H Lin (eds), New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers. Communications in Computer and Information Science, vol. 1013, Springer Verlag, pp. 681-687, 23rd International Computer Symposium, ICS 2018, Yunlin, Taiwan, 2018/12/20. https://doi.org/10.1007/978-981-13-9190-3_75

Yeh CH, Huang YP, Chen MJ. Scale Invariant Multi-view Depth Estimation Network with cGAN Refinement. In Chang CY, Lin CC, Lin HH, editors, New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers. Springer Verlag. 2019. p. 681-687. (Communications in Computer and Information Science). doi: 10.1007/978-981-13-9190-3_75

Yeh, Chia Hung ; Huang, Yao Pao ; Chen, Mei Juan. / Scale Invariant Multi-view Depth Estimation Network with cGAN Refinement. New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers. editor / Chuan-Yu Chang ; Chien-Chou Lin ; Horng-Horng Lin. Springer Verlag, 2019. pp. 681-687 (Communications in Computer and Information Science).

@inproceedings{b712235f57284213960267f605d2eb09,

title = "Scale Invariant Multi-view Depth Estimation Network with cGAN Refinement",

abstract = "In this paper we propose a deep learning based depth estimation method for monocular RGB sequences. We train a pair of encoder-decoder network to resolve depth information form image pairs and relative camera poses. To solve scale ambiguous of monocular sequences, a conditional generative adversarial network is applied. Experimental results show that the proposed method can overcome the problem of scale ambiguous and therefore is more suitable for a variety of applications.",

keywords = "Conditional generative adversarial network, Deep learning, Multi-view depth estimation",

author = "Yeh, {Chia Hung} and Huang, {Yao Pao} and Chen, {Mei Juan}",

note = "Publisher Copyright: {\textcopyright} Springer Nature Singapore Pte Ltd. 2019.; 23rd International Computer Symposium, ICS 2018 ; Conference date: 20-12-2018 Through 22-12-2018",

year = "2019",

doi = "10.1007/978-981-13-9190-3_75",

language = "English",

isbn = "9789811391897",

series = "Communications in Computer and Information Science",

publisher = "Springer Verlag",

pages = "681--687",

editor = "Chuan-Yu Chang and Chien-Chou Lin and Horng-Horng Lin",

booktitle = "New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers",

}

TY - GEN

T1 - Scale Invariant Multi-view Depth Estimation Network with cGAN Refinement

AU - Yeh, Chia Hung

AU - Huang, Yao Pao

AU - Chen, Mei Juan

N1 - Publisher Copyright: © Springer Nature Singapore Pte Ltd. 2019.

PY - 2019

Y1 - 2019

N2 - In this paper we propose a deep learning based depth estimation method for monocular RGB sequences. We train a pair of encoder-decoder network to resolve depth information form image pairs and relative camera poses. To solve scale ambiguous of monocular sequences, a conditional generative adversarial network is applied. Experimental results show that the proposed method can overcome the problem of scale ambiguous and therefore is more suitable for a variety of applications.

AB - In this paper we propose a deep learning based depth estimation method for monocular RGB sequences. We train a pair of encoder-decoder network to resolve depth information form image pairs and relative camera poses. To solve scale ambiguous of monocular sequences, a conditional generative adversarial network is applied. Experimental results show that the proposed method can overcome the problem of scale ambiguous and therefore is more suitable for a variety of applications.

KW - Conditional generative adversarial network

KW - Deep learning

KW - Multi-view depth estimation

UR - http://www.scopus.com/inward/record.url?scp=85069739047&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069739047&partnerID=8YFLogxK

U2 - 10.1007/978-981-13-9190-3_75

DO - 10.1007/978-981-13-9190-3_75

M3 - Conference contribution

AN - SCOPUS:85069739047

SN - 9789811391897

T3 - Communications in Computer and Information Science

SP - 681

EP - 687

BT - New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers

A2 - Chang, Chuan-Yu

A2 - Lin, Chien-Chou

A2 - Lin, Horng-Horng

PB - Springer Verlag

T2 - 23rd International Computer Symposium, ICS 2018

Y2 - 20 December 2018 through 22 December 2018

ER -

Scale Invariant Multi-view Depth Estimation Network with cGAN Refinement

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this