Perceptual Quality Study on Deep Learning Based Image Compression

Zhengxue Cheng; Pinar Akyazi; Heming Sun; Jiro Katto; Touradj Ebrahimi

doi:10.1109/ICIP.2019.8803824

Perceptual Quality Study on Deep Learning Based Image Compression

Zhengxue Cheng, Pinar Akyazi, Heming Sun, Jiro Katto, Touradj Ebrahimi

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

1 Citation (Scopus)

Abstract

Recently deep learning based image compression has made rapid advances with promising results based on objective quality metrics. However, a rigorous subjective quality evaluation on such compression schemes have rarely been reported. This paper aims at perceptual quality studies on learned compression. First, we build a general learned compression approach, and optimize the model. In total six compression algorithms are considered for this study. Then, we perform subjective quality tests in a controlled environment using high-resolution images. Results demonstrate learned compression optimized by MS-SSIM yields competitive results that approach the efficiency of state-of-the-art compression. The results obtained can provide a useful benchmark for future developments in learned image compression.

Original language	English
Title of host publication	2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings
Publisher	IEEE Computer Society
Pages	719-723
Number of pages	5
ISBN (Electronic)	9781538662496
DOIs	https://doi.org/10.1109/ICIP.2019.8803824
Publication status	Published - 2019 Sep
Event	26th IEEE International Conference on Image Processing, ICIP 2019 - Taipei, Taiwan, Province of China Duration: 2019 Sep 22 → 2019 Sep 25

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
Volume	2019-September
ISSN (Print)	1522-4880

Conference

Conference	26th IEEE International Conference on Image Processing, ICIP 2019
Country	Taiwan, Province of China
City	Taipei
Period	19/9/22 → 19/9/25

Keywords

Subjective and objective quality evaluation
compression standards
learning image compression

ASJC Scopus subject areas

Software
Computer Vision and Pattern Recognition
Signal Processing

Access to Document

10.1109/ICIP.2019.8803824

Fingerprint Dive into the research topics of 'Perceptual Quality Study on Deep Learning Based Image Compression'. Together they form a unique fingerprint.

Cite this

Cheng, Z., Akyazi, P., Sun, H., Katto, J., & Ebrahimi, T. (2019). Perceptual Quality Study on Deep Learning Based Image Compression. In 2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings (pp. 719-723). [8803824] (Proceedings - International Conference on Image Processing, ICIP; Vol. 2019-September). IEEE Computer Society. https://doi.org/10.1109/ICIP.2019.8803824

Perceptual Quality Study on Deep Learning Based Image Compression. / Cheng, Zhengxue; Akyazi, Pinar; Sun, Heming ; Katto, Jiro; Ebrahimi, Touradj.

2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings. IEEE Computer Society, 2019. p. 719-723 8803824 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2019-September).

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

Cheng, Z, Akyazi, P, Sun, H , Katto, J & Ebrahimi, T 2019, Perceptual Quality Study on Deep Learning Based Image Compression. in 2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings., 8803824, Proceedings - International Conference on Image Processing, ICIP, vol. 2019-September, IEEE Computer Society, pp. 719-723, 26th IEEE International Conference on Image Processing, ICIP 2019, Taipei, Taiwan, Province of China, 19/9/22. https://doi.org/10.1109/ICIP.2019.8803824

@inproceedings{c5287f37e504433585708e79767272f6,

title = "Perceptual Quality Study on Deep Learning Based Image Compression",

abstract = "Recently deep learning based image compression has made rapid advances with promising results based on objective quality metrics. However, a rigorous subjective quality evaluation on such compression schemes have rarely been reported. This paper aims at perceptual quality studies on learned compression. First, we build a general learned compression approach, and optimize the model. In total six compression algorithms are considered for this study. Then, we perform subjective quality tests in a controlled environment using high-resolution images. Results demonstrate learned compression optimized by MS-SSIM yields competitive results that approach the efficiency of state-of-the-art compression. The results obtained can provide a useful benchmark for future developments in learned image compression.",

keywords = "Subjective and objective quality evaluation, compression standards, learning image compression",

author = "Zhengxue Cheng and Pinar Akyazi and Heming Sun and Jiro Katto and Touradj Ebrahimi",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.; 26th IEEE International Conference on Image Processing, ICIP 2019 ; Conference date: 22-09-2019 Through 25-09-2019",

year = "2019",

month = sep,

doi = "10.1109/ICIP.2019.8803824",

language = "English",

series = "Proceedings - International Conference on Image Processing, ICIP",

publisher = "IEEE Computer Society",

pages = "719--723",

booktitle = "2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings",

}

TY - GEN

T1 - Perceptual Quality Study on Deep Learning Based Image Compression

AU - Cheng, Zhengxue

AU - Akyazi, Pinar

AU - Sun, Heming

AU - Katto, Jiro

AU - Ebrahimi, Touradj

PY - 2019/9

Y1 - 2019/9

N2 - Recently deep learning based image compression has made rapid advances with promising results based on objective quality metrics. However, a rigorous subjective quality evaluation on such compression schemes have rarely been reported. This paper aims at perceptual quality studies on learned compression. First, we build a general learned compression approach, and optimize the model. In total six compression algorithms are considered for this study. Then, we perform subjective quality tests in a controlled environment using high-resolution images. Results demonstrate learned compression optimized by MS-SSIM yields competitive results that approach the efficiency of state-of-the-art compression. The results obtained can provide a useful benchmark for future developments in learned image compression.

AB - Recently deep learning based image compression has made rapid advances with promising results based on objective quality metrics. However, a rigorous subjective quality evaluation on such compression schemes have rarely been reported. This paper aims at perceptual quality studies on learned compression. First, we build a general learned compression approach, and optimize the model. In total six compression algorithms are considered for this study. Then, we perform subjective quality tests in a controlled environment using high-resolution images. Results demonstrate learned compression optimized by MS-SSIM yields competitive results that approach the efficiency of state-of-the-art compression. The results obtained can provide a useful benchmark for future developments in learned image compression.

KW - Subjective and objective quality evaluation

KW - compression standards

KW - learning image compression

UR - http://www.scopus.com/inward/record.url?scp=85076800720&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85076800720&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2019.8803824

DO - 10.1109/ICIP.2019.8803824

M3 - Conference contribution

AN - SCOPUS:85076800720

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 719

EP - 723

BT - 2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings

PB - IEEE Computer Society

T2 - 26th IEEE International Conference on Image Processing, ICIP 2019

Y2 - 22 September 2019 through 25 September 2019

ER -