Siralama Sorunu Olarak Nispi Derinlik Tahmini Relative Depth Estimation as a Ranking Problem

Alican Mertan; Damien Jade Duff; Gozde Unal

doi:10.1109/SIU49456.2020.9302314

Siralama Sorunu Olarak Nispi Derinlik Tahmini Relative Depth Estimation as a Ranking Problem

Alican Mertan, Damien Jade Duff, Gozde Unal

Department of Artificial Intelligence and Data Engineering

Istanbul Technical University

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

Abstract

We present a formulation of the relative depth estimation from a single image problem, as a ranking problem. By reformulating the problem this way, we were able to utilize literature on the ranking problem, and apply the existing knowledge to achieve better results. To this end, we have introduced a listwise ranking loss borrowed from ranking literature, weighted ListMLE, to the relative depth estimation problem. We have also brought a new metric which considers pixel depth ranking accuracy, on which our method is stronger.

Original language	English
Title of host publication	2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728172064
DOIs	https://doi.org/10.1109/SIU49456.2020.9302314
Publication status	Published - 5 Oct 2020
Event	28th Signal Processing and Communications Applications Conference, SIU 2020 - Gaziantep, Turkey Duration: 5 Oct 2020 → 7 Oct 2020

Publication series

Name	2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings

Conference

Conference	28th Signal Processing and Communications Applications Conference, SIU 2020
Country/Territory	Turkey
City	Gaziantep
Period	5/10/20 → 7/10/20

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

Funding

The work is supported by the Scientific and Technological Research Council of Turkey (TÜBITAK), project 116E167.

Funders	Funder number
Türkiye Bilimsel ve Teknolojik Araştirma Kurumu	116E167

Keywords

deep learning
depth estimation
learning to rank
relative depth estimation

Access to Document

10.1109/SIU49456.2020.9302314

Cite this

Mertan, A., Duff, D. J., & Unal, G. (2020). Siralama Sorunu Olarak Nispi Derinlik Tahmini Relative Depth Estimation as a Ranking Problem. In 2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings Article 9302314 (2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SIU49456.2020.9302314

Mertan, Alican ; Duff, Damien Jade ; Unal, Gozde. / Siralama Sorunu Olarak Nispi Derinlik Tahmini Relative Depth Estimation as a Ranking Problem. 2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. (2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings).

@inproceedings{5d38487b5a4344a287e8e789c0b5c038,

title = "Siralama Sorunu Olarak Nispi Derinlik Tahmini Relative Depth Estimation as a Ranking Problem",

abstract = "We present a formulation of the relative depth estimation from a single image problem, as a ranking problem. By reformulating the problem this way, we were able to utilize literature on the ranking problem, and apply the existing knowledge to achieve better results. To this end, we have introduced a listwise ranking loss borrowed from ranking literature, weighted ListMLE, to the relative depth estimation problem. We have also brought a new metric which considers pixel depth ranking accuracy, on which our method is stronger.",

keywords = "deep learning, depth estimation, learning to rank, relative depth estimation",

author = "Alican Mertan and Duff, {Damien Jade} and Gozde Unal",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 28th Signal Processing and Communications Applications Conference, SIU 2020 ; Conference date: 05-10-2020 Through 07-10-2020",

year = "2020",

month = oct,

day = "5",

doi = "10.1109/SIU49456.2020.9302314",

language = "English",

series = "2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings",

address = "United States",

}

Mertan, A, Duff, DJ & Unal, G 2020, Siralama Sorunu Olarak Nispi Derinlik Tahmini Relative Depth Estimation as a Ranking Problem. in 2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings., 9302314, 2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings, Institute of Electrical and Electronics Engineers Inc., 28th Signal Processing and Communications Applications Conference, SIU 2020, Gaziantep, Turkey, 5/10/20. https://doi.org/10.1109/SIU49456.2020.9302314

Siralama Sorunu Olarak Nispi Derinlik Tahmini Relative Depth Estimation as a Ranking Problem. / Mertan, Alican; Duff, Damien Jade; Unal, Gozde.
2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. 9302314 (2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings).

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

TY - GEN

T1 - Siralama Sorunu Olarak Nispi Derinlik Tahmini Relative Depth Estimation as a Ranking Problem

AU - Mertan, Alican

AU - Duff, Damien Jade

AU - Unal, Gozde

PY - 2020/10/5

Y1 - 2020/10/5

N2 - We present a formulation of the relative depth estimation from a single image problem, as a ranking problem. By reformulating the problem this way, we were able to utilize literature on the ranking problem, and apply the existing knowledge to achieve better results. To this end, we have introduced a listwise ranking loss borrowed from ranking literature, weighted ListMLE, to the relative depth estimation problem. We have also brought a new metric which considers pixel depth ranking accuracy, on which our method is stronger.

AB - We present a formulation of the relative depth estimation from a single image problem, as a ranking problem. By reformulating the problem this way, we were able to utilize literature on the ranking problem, and apply the existing knowledge to achieve better results. To this end, we have introduced a listwise ranking loss borrowed from ranking literature, weighted ListMLE, to the relative depth estimation problem. We have also brought a new metric which considers pixel depth ranking accuracy, on which our method is stronger.

KW - deep learning

KW - depth estimation

KW - learning to rank

KW - relative depth estimation

UR - http://www.scopus.com/inward/record.url?scp=85100294326&partnerID=8YFLogxK

U2 - 10.1109/SIU49456.2020.9302314

DO - 10.1109/SIU49456.2020.9302314

M3 - Conference contribution

AN - SCOPUS:85100294326

T3 - 2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings

BT - 2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 28th Signal Processing and Communications Applications Conference, SIU 2020

Y2 - 5 October 2020 through 7 October 2020

ER -

Mertan A, Duff DJ, Unal G. Siralama Sorunu Olarak Nispi Derinlik Tahmini Relative Depth Estimation as a Ranking Problem. In 2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2020. 9302314. (2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings). doi: 10.1109/SIU49456.2020.9302314

Siralama Sorunu Olarak Nispi Derinlik Tahmini Relative Depth Estimation as a Ranking Problem

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this