NTIRE 2021 challenge for defocus deblurring using dual-pixel images: Methods and results

NTIRE 2021

doi:10.1109/CVPRW53098.2021.00070

NTIRE 2021 challenge for defocus deblurring using dual-pixel images: Methods and results

NTIRE 2021

Research output: A Conference proceeding or a Chapter in Book › Conference contribution › peer-review

24 Citations (Scopus)

Abstract

This paper provides a review of the NTIRE 2021 challenge targeting defocus deblurring using dual-pixel (DP) data. The goal of this single-track challenge was to reduce spatially varying defocus blur present in images captured with a shallow depth of field. The images used in this challenge were obtained using a DP sensor that provided a pair of DP views per captured image. Submitted solutions were evaluated using conventional signal processing metrics, namely peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). Out of 185 registered participants, nine teams provided methods and competed in the final stage. The paper describes the methods proposed by the participating teams and their results. The winning teams represent the state-of-the-art in terms of defocus de-blurring using DP images.

Original language	English
Title of host publication	Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021
Editors	David Forsyth, Georgia Gkioxari, Tinne Tuytelaars, Ruigang Yang, Jingyi Yu, Michael S. Brown, Rahul Sukthankar, Tieniu Tan, Lihi Zelnik
Place of Publication	United States
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	578-587
Number of pages	10
ISBN (Electronic)	9781665448994
ISBN (Print)	9781665449007
DOIs	https://doi.org/10.1109/CVPRW53098.2021.00070
Publication status	Published - Jun 2021
Externally published	Yes
Event	2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021 - Virtual, Online, United States Duration: 19 Jun 2021 → 25 Jun 2021

Publication series

Name	IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
ISSN (Print)	2160-7508
ISSN (Electronic)	2160-7516

Conference

Conference	2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021
Country/Territory	United States
City	Virtual, Online
Period	19/06/21 → 25/06/21

Access to Document

10.1109/CVPRW53098.2021.00070

Cite this

NTIRE 2021 (2021). NTIRE 2021 challenge for defocus deblurring using dual-pixel images: Methods and results. In D. Forsyth, G. Gkioxari, T. Tuytelaars, R. Yang, J. Yu, M. S. Brown, R. Sukthankar, T. Tan, & L. Zelnik (Eds.), Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021 (pp. 578-587). (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops). IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/CVPRW53098.2021.00070

NTIRE 2021. / NTIRE 2021 challenge for defocus deblurring using dual-pixel images : Methods and results. Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021. editor / David Forsyth ; Georgia Gkioxari ; Tinne Tuytelaars ; Ruigang Yang ; Jingyi Yu ; Michael S. Brown ; Rahul Sukthankar ; Tieniu Tan ; Lihi Zelnik. United States : IEEE, Institute of Electrical and Electronics Engineers, 2021. pp. 578-587 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops).

@inproceedings{f5ec1fb03cec4c23a7c7d6774218eea3,

title = "NTIRE 2021 challenge for defocus deblurring using dual-pixel images: Methods and results",

abstract = "This paper provides a review of the NTIRE 2021 challenge targeting defocus deblurring using dual-pixel (DP) data. The goal of this single-track challenge was to reduce spatially varying defocus blur present in images captured with a shallow depth of field. The images used in this challenge were obtained using a DP sensor that provided a pair of DP views per captured image. Submitted solutions were evaluated using conventional signal processing metrics, namely peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). Out of 185 registered participants, nine teams provided methods and competed in the final stage. The paper describes the methods proposed by the participating teams and their results. The winning teams represent the state-of-the-art in terms of defocus de-blurring using DP images.",

author = "{NTIRE 2021} and Abdullah Abuolaim and Radu Timofte and Brown, {Michael S.} and Dafeng Zhang and Xiaobing Wang and Zamir, {Syed Waqas} and Aditya Arora and Salman Khan and Munawar Hayat and Khan, {Fahad Shahbaz} and Ling Shao and Shuai Liu",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021 ; Conference date: 19-06-2021 Through 25-06-2021",

year = "2021",

month = jun,

doi = "10.1109/CVPRW53098.2021.00070",

language = "English",

isbn = "9781665449007",

series = "IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

pages = "578--587",

editor = "David Forsyth and Georgia Gkioxari and Tinne Tuytelaars and Ruigang Yang and Jingyi Yu and Brown, {Michael S.} and Rahul Sukthankar and Tieniu Tan and Lihi Zelnik",

booktitle = "Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021",

address = "United States",

}

NTIRE 2021 2021, NTIRE 2021 challenge for defocus deblurring using dual-pixel images: Methods and results. in D Forsyth, G Gkioxari, T Tuytelaars, R Yang, J Yu, MS Brown, R Sukthankar, T Tan & L Zelnik (eds), Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021. IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, IEEE, Institute of Electrical and Electronics Engineers, United States, pp. 578-587, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021, Virtual, Online, United States, 19/06/21. https://doi.org/10.1109/CVPRW53098.2021.00070

NTIRE 2021 challenge for defocus deblurring using dual-pixel images: Methods and results. / NTIRE 2021.
Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021. ed. / David Forsyth; Georgia Gkioxari; Tinne Tuytelaars; Ruigang Yang; Jingyi Yu; Michael S. Brown; Rahul Sukthankar; Tieniu Tan; Lihi Zelnik. United States: IEEE, Institute of Electrical and Electronics Engineers, 2021. p. 578-587 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops).

Research output: A Conference proceeding or a Chapter in Book › Conference contribution › peer-review

TY - GEN

T1 - NTIRE 2021 challenge for defocus deblurring using dual-pixel images

T2 - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021

AU - NTIRE 2021

AU - Abuolaim, Abdullah

AU - Timofte, Radu

AU - Brown, Michael S.

AU - Zhang, Dafeng

AU - Wang, Xiaobing

AU - Zamir, Syed Waqas

AU - Arora, Aditya

AU - Khan, Salman

AU - Hayat, Munawar

AU - Khan, Fahad Shahbaz

AU - Shao, Ling

AU - Liu, Shuai

PY - 2021/6

Y1 - 2021/6

N2 - This paper provides a review of the NTIRE 2021 challenge targeting defocus deblurring using dual-pixel (DP) data. The goal of this single-track challenge was to reduce spatially varying defocus blur present in images captured with a shallow depth of field. The images used in this challenge were obtained using a DP sensor that provided a pair of DP views per captured image. Submitted solutions were evaluated using conventional signal processing metrics, namely peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). Out of 185 registered participants, nine teams provided methods and competed in the final stage. The paper describes the methods proposed by the participating teams and their results. The winning teams represent the state-of-the-art in terms of defocus de-blurring using DP images.

AB - This paper provides a review of the NTIRE 2021 challenge targeting defocus deblurring using dual-pixel (DP) data. The goal of this single-track challenge was to reduce spatially varying defocus blur present in images captured with a shallow depth of field. The images used in this challenge were obtained using a DP sensor that provided a pair of DP views per captured image. Submitted solutions were evaluated using conventional signal processing metrics, namely peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). Out of 185 registered participants, nine teams provided methods and competed in the final stage. The paper describes the methods proposed by the participating teams and their results. The winning teams represent the state-of-the-art in terms of defocus de-blurring using DP images.

UR - http://www.scopus.com/inward/record.url?scp=85114851721&partnerID=8YFLogxK

U2 - 10.1109/CVPRW53098.2021.00070

DO - 10.1109/CVPRW53098.2021.00070

M3 - Conference contribution

AN - SCOPUS:85114851721

SN - 9781665449007

T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops

SP - 578

EP - 587

BT - Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021

A2 - Forsyth, David

A2 - Gkioxari, Georgia

A2 - Tuytelaars, Tinne

A2 - Yang, Ruigang

A2 - Yu, Jingyi

A2 - Brown, Michael S.

A2 - Sukthankar, Rahul

A2 - Tan, Tieniu

A2 - Zelnik, Lihi

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - United States

Y2 - 19 June 2021 through 25 June 2021

ER -

NTIRE 2021. NTIRE 2021 challenge for defocus deblurring using dual-pixel images: Methods and results. In Forsyth D, Gkioxari G, Tuytelaars T, Yang R, Yu J, Brown MS, Sukthankar R, Tan T, Zelnik L, editors, Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021. United States: IEEE, Institute of Electrical and Electronics Engineers. 2021. p. 578-587. (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops). doi: 10.1109/CVPRW53098.2021.00070

NTIRE 2021 challenge for defocus deblurring using dual-pixel images: Methods and results

Abstract

Publication series

Conference

Access to Document

Other files and links

Cite this