Image Restoration with Deep Generative Models

Raymond A. Yeh, Teck Yian Lim, Chen Chen, Alexander Gerhard Schwing, Mark Allan Hasegawa-Johnson, Minh N Do

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Many image restoration problems are ill-posed in nature, hence, beyond the input image, most existing methods rely on a carefully engineered image prior, which enforces some local image consistency in the recovered image. How tightly the prior assumptions are fulfilled has a big impact on the resulting task performance. To obtain more flexibility, in this work, we proposed to design the image prior in a data-driven manner. Instead of explicitly defining the prior, we learn it using deep generative models. We demonstrate that this learned prior can be applied to many image restoration problems using an unified framework.

Original languageEnglish (US)
Title of host publication2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages6772-6776
Number of pages5
ISBN (Print)9781538646588
DOIs
StatePublished - Sep 10 2018
Event2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Calgary, Canada
Duration: Apr 15 2018Apr 20 2018

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2018-April
ISSN (Print)1520-6149

Other

Other2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018
CountryCanada
CityCalgary
Period4/15/184/20/18

Fingerprint

Image reconstruction

Keywords

  • Deep generative models
  • Generative adversarial networks
  • Image restoration

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Yeh, R. A., Lim, T. Y., Chen, C., Schwing, A. G., Hasegawa-Johnson, M. A., & Do, M. N. (2018). Image Restoration with Deep Generative Models. In 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings (pp. 6772-6776). [8462317] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2018-April). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2018.8462317

Image Restoration with Deep Generative Models. / Yeh, Raymond A.; Lim, Teck Yian; Chen, Chen; Schwing, Alexander Gerhard; Hasegawa-Johnson, Mark Allan; Do, Minh N.

2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. p. 6772-6776 8462317 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2018-April).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yeh, RA, Lim, TY, Chen, C, Schwing, AG, Hasegawa-Johnson, MA & Do, MN 2018, Image Restoration with Deep Generative Models. in 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings., 8462317, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2018-April, Institute of Electrical and Electronics Engineers Inc., pp. 6772-6776, 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018, Calgary, Canada, 4/15/18. https://doi.org/10.1109/ICASSP.2018.8462317
Yeh RA, Lim TY, Chen C, Schwing AG, Hasegawa-Johnson MA, Do MN. Image Restoration with Deep Generative Models. In 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2018. p. 6772-6776. 8462317. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2018.8462317
Yeh, Raymond A. ; Lim, Teck Yian ; Chen, Chen ; Schwing, Alexander Gerhard ; Hasegawa-Johnson, Mark Allan ; Do, Minh N. / Image Restoration with Deep Generative Models. 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 6772-6776 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).
@inproceedings{ae0fc58d24584ca3ba6e711cb39aaf66,
title = "Image Restoration with Deep Generative Models",
abstract = "Many image restoration problems are ill-posed in nature, hence, beyond the input image, most existing methods rely on a carefully engineered image prior, which enforces some local image consistency in the recovered image. How tightly the prior assumptions are fulfilled has a big impact on the resulting task performance. To obtain more flexibility, in this work, we proposed to design the image prior in a data-driven manner. Instead of explicitly defining the prior, we learn it using deep generative models. We demonstrate that this learned prior can be applied to many image restoration problems using an unified framework.",
keywords = "Deep generative models, Generative adversarial networks, Image restoration",
author = "Yeh, {Raymond A.} and Lim, {Teck Yian} and Chen Chen and Schwing, {Alexander Gerhard} and Hasegawa-Johnson, {Mark Allan} and Do, {Minh N}",
year = "2018",
month = "9",
day = "10",
doi = "10.1109/ICASSP.2018.8462317",
language = "English (US)",
isbn = "9781538646588",
series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "6772--6776",
booktitle = "2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings",
address = "United States",

}

TY - GEN

T1 - Image Restoration with Deep Generative Models

AU - Yeh, Raymond A.

AU - Lim, Teck Yian

AU - Chen, Chen

AU - Schwing, Alexander Gerhard

AU - Hasegawa-Johnson, Mark Allan

AU - Do, Minh N

PY - 2018/9/10

Y1 - 2018/9/10

N2 - Many image restoration problems are ill-posed in nature, hence, beyond the input image, most existing methods rely on a carefully engineered image prior, which enforces some local image consistency in the recovered image. How tightly the prior assumptions are fulfilled has a big impact on the resulting task performance. To obtain more flexibility, in this work, we proposed to design the image prior in a data-driven manner. Instead of explicitly defining the prior, we learn it using deep generative models. We demonstrate that this learned prior can be applied to many image restoration problems using an unified framework.

AB - Many image restoration problems are ill-posed in nature, hence, beyond the input image, most existing methods rely on a carefully engineered image prior, which enforces some local image consistency in the recovered image. How tightly the prior assumptions are fulfilled has a big impact on the resulting task performance. To obtain more flexibility, in this work, we proposed to design the image prior in a data-driven manner. Instead of explicitly defining the prior, we learn it using deep generative models. We demonstrate that this learned prior can be applied to many image restoration problems using an unified framework.

KW - Deep generative models

KW - Generative adversarial networks

KW - Image restoration

UR - http://www.scopus.com/inward/record.url?scp=85054209504&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054209504&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2018.8462317

DO - 10.1109/ICASSP.2018.8462317

M3 - Conference contribution

AN - SCOPUS:85054209504

SN - 9781538646588

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 6772

EP - 6776

BT - 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -