Deep blind image quality assessment by learning sensitivity map

Jongyoo Kim; Woojae Kim; Sanghoon Lee

doi:10.1109/ICASSP.2018.8462369

Deep blind image quality assessment by learning sensitivity map

Jongyoo Kim, Woojae Kim, Sanghoon Lee

Department of Electrical and Electronic Engineering

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

4 Citations (Scopus)

Abstract

Applying a deep convolutional neural network CNN to no-reference image quality assessment (NR-IQA) is a challenging task due to the lack of a training database. In this paper, we propose a CNN-based NR-IQA framework that can effectively solve this problem. The proposed method-the Deep Blind image Quality Assessment predictor (DeepBQA)-adopts two step training stages to avoid overfitting. In the first stage, a ground-truth objective error map is generated and used as a proxy training target. Then, in the second stage, subjective score is predicted by learning a sensitivity map, which weights each pixel in the predicted objective error map. To compensate the inaccurate prediction of the objective error on the homogeneous regions, we additionally suggest a reliability map. Experiments showed that DeepBQA yields a state-of-the-art correlation with human opinions.

Original language	English
Title of host publication	2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	6727-6731
Number of pages	5
ISBN (Print)	9781538646588
DOIs	https://doi.org/10.1109/ICASSP.2018.8462369
Publication status	Published - 2018 Sept 10
Event	2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Calgary, Canada Duration: 2018 Apr 15 → 2018 Apr 20

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2018-April
ISSN (Print)	1520-6149

Conference

Conference	2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018
Country/Territory	Canada
City	Calgary
Period	18/4/15 → 18/4/20

Bibliographical note

Funding Information:
This work is done when Jongyoo Kim is an intern at Microsoft Research Asia This work was supported by Samsung Research Funding Center of Sam-sung Electronics under Project Number SRFC-IT1702-08

Publisher Copyright:
© 2018 IEEE.

All Science Journal Classification (ASJC) codes

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP.2018.8462369

Cite this

Kim, J., Kim, W., & Lee, S. (2018). Deep blind image quality assessment by learning sensitivity map. In 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings (pp. 6727-6731). Article 8462369 (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.2018.8462369

Kim, Jongyoo ; Kim, Woojae ; Lee, Sanghoon. / Deep blind image quality assessment by learning sensitivity map. 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 6727-6731 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{c069969d7d704396a6891f0f0f6dc25a,

title = "Deep blind image quality assessment by learning sensitivity map",

abstract = "Applying a deep convolutional neural network CNN to no-reference image quality assessment (NR-IQA) is a challenging task due to the lack of a training database. In this paper, we propose a CNN-based NR-IQA framework that can effectively solve this problem. The proposed method-the Deep Blind image Quality Assessment predictor (DeepBQA)-adopts two step training stages to avoid overfitting. In the first stage, a ground-truth objective error map is generated and used as a proxy training target. Then, in the second stage, subjective score is predicted by learning a sensitivity map, which weights each pixel in the predicted objective error map. To compensate the inaccurate prediction of the objective error on the homogeneous regions, we additionally suggest a reliability map. Experiments showed that DeepBQA yields a state-of-the-art correlation with human opinions.",

author = "Jongyoo Kim and Woojae Kim and Sanghoon Lee",

note = "Funding Information: This work is done when Jongyoo Kim is an intern at Microsoft Research Asia This work was supported by Samsung Research Funding Center of Sam-sung Electronics under Project Number SRFC-IT1702-08 Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 ; Conference date: 15-04-2018 Through 20-04-2018",

year = "2018",

month = sep,

day = "10",

doi = "10.1109/ICASSP.2018.8462369",

language = "English",

isbn = "9781538646588",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

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

pages = "6727--6731",

booktitle = "2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings",

address = "United States",

}

Kim, J, Kim, W & Lee, S 2018, Deep blind image quality assessment by learning sensitivity map. in 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings., 8462369, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2018-April, Institute of Electrical and Electronics Engineers Inc., pp. 6727-6731, 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018, Calgary, Canada, 18/4/15. https://doi.org/10.1109/ICASSP.2018.8462369

Deep blind image quality assessment by learning sensitivity map. / Kim, Jongyoo; Kim, Woojae; Lee, Sanghoon.
2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. p. 6727-6731 8462369 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2018-April).

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

TY - GEN

T1 - Deep blind image quality assessment by learning sensitivity map

AU - Kim, Jongyoo

AU - Kim, Woojae

AU - Lee, Sanghoon

N1 - Funding Information: This work is done when Jongyoo Kim is an intern at Microsoft Research Asia This work was supported by Samsung Research Funding Center of Sam-sung Electronics under Project Number SRFC-IT1702-08 Publisher Copyright: © 2018 IEEE.

PY - 2018/9/10

Y1 - 2018/9/10

N2 - Applying a deep convolutional neural network CNN to no-reference image quality assessment (NR-IQA) is a challenging task due to the lack of a training database. In this paper, we propose a CNN-based NR-IQA framework that can effectively solve this problem. The proposed method-the Deep Blind image Quality Assessment predictor (DeepBQA)-adopts two step training stages to avoid overfitting. In the first stage, a ground-truth objective error map is generated and used as a proxy training target. Then, in the second stage, subjective score is predicted by learning a sensitivity map, which weights each pixel in the predicted objective error map. To compensate the inaccurate prediction of the objective error on the homogeneous regions, we additionally suggest a reliability map. Experiments showed that DeepBQA yields a state-of-the-art correlation with human opinions.

AB - Applying a deep convolutional neural network CNN to no-reference image quality assessment (NR-IQA) is a challenging task due to the lack of a training database. In this paper, we propose a CNN-based NR-IQA framework that can effectively solve this problem. The proposed method-the Deep Blind image Quality Assessment predictor (DeepBQA)-adopts two step training stages to avoid overfitting. In the first stage, a ground-truth objective error map is generated and used as a proxy training target. Then, in the second stage, subjective score is predicted by learning a sensitivity map, which weights each pixel in the predicted objective error map. To compensate the inaccurate prediction of the objective error on the homogeneous regions, we additionally suggest a reliability map. Experiments showed that DeepBQA yields a state-of-the-art correlation with human opinions.

UR - http://www.scopus.com/inward/record.url?scp=85054226748&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054226748&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2018.8462369

DO - 10.1109/ICASSP.2018.8462369

M3 - Conference contribution

AN - SCOPUS:85054226748

SN - 9781538646588

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

SP - 6727

EP - 6731

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

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018

Y2 - 15 April 2018 through 20 April 2018

ER -

Kim J, Kim W, Lee S. Deep blind image quality assessment by learning sensitivity map. In 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2018. p. 6727-6731. 8462369. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.2018.8462369

Deep blind image quality assessment by learning sensitivity map

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this