DRASIC: Distributed recurrent autoencoder for scalable image compression

Enmao Diao; Jie Ding; Vahid Tarokh

doi:10.1109/DCC47342.2020.00008

DRASIC: Distributed recurrent autoencoder for scalable image compression

Enmao Diao, Jie Ding, Vahid Tarokh

Statistics (Twin Cities)

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

11 Scopus citations

Abstract

We propose a new architecture for distributed image compression from a group of distributed data sources. The work is motivated by practical needs of data-driven codec design, low power consumption, robustness, and data privacy. The proposed architecture, which we refer to as Distributed Recurrent Autoencoder for Scalable Image Compression (DRASIC), is able to train distributed encoders and one joint decoder on correlated data sources. Its compression capability is much better than the method of training codecs separately. Meanwhile, the performance of our distributed system with 10 distributed sources is only within 2 dB peak signal-to-noise ratio (PSNR) of the performance of a single codec trained with all data sources. We experiment distributed sources with different correlations and show how our data-driven methodology well matches the Slepian-Wolf Theorem in Distributed Source Coding (DSC). To the best of our knowledge, this is the first data-driven DSC framework for general distributed code design with deep learning.

Original language	English (US)
Title of host publication	Proceedings - DCC 2020
Subtitle of host publication	Data Compression Conference
Editors	Ali Bilgin, Michael W. Marcellin, Joan Serra-Sagrista, James A. Storer
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	3-12
Number of pages	10
ISBN (Electronic)	9781728164571
DOIs	https://doi.org/10.1109/DCC47342.2020.00008
State	Published - Mar 2020
Event	2020 Data Compression Conference, DCC 2020 - Snowbird, United States Duration: Mar 24 2020 → Mar 27 2020

Publication series

Name	Data Compression Conference Proceedings
Volume	2020-March
ISSN (Print)	1068-0314

Conference

Conference	2020 Data Compression Conference, DCC 2020
Country/Territory	United States
City	Snowbird
Period	3/24/20 → 3/27/20

Bibliographical note

Funding Information:
This work was supported in part by Office of Naval Research Grant No. N00014-18-1-2244. We provide our implementation at https://github.com/dem123456789/ Distributed-Recurrent-Autoencoder-for-Scalable-Image-Compression

Publisher Copyright:
© 2020 IEEE.

Access

10.1109/DCC47342.2020.00008

OpenUrl availability

Full text

Cite this

Diao, E., Ding, J., & Tarokh, V. (2020). DRASIC: Distributed recurrent autoencoder for scalable image compression. In A. Bilgin, M. W. Marcellin, J. Serra-Sagrista, & J. A. Storer (Eds.), Proceedings - DCC 2020: Data Compression Conference (pp. 3-12). Article 9105700 (Data Compression Conference Proceedings; Vol. 2020-March). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/DCC47342.2020.00008

DRASIC: Distributed recurrent autoencoder for scalable image compression. / Diao, Enmao; Ding, Jie; Tarokh, Vahid.
Proceedings - DCC 2020: Data Compression Conference. ed. / Ali Bilgin; Michael W. Marcellin; Joan Serra-Sagrista; James A. Storer. Institute of Electrical and Electronics Engineers Inc., 2020. p. 3-12 9105700 (Data Compression Conference Proceedings; Vol. 2020-March).

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

Diao, E, Ding, J & Tarokh, V 2020, DRASIC: Distributed recurrent autoencoder for scalable image compression. in A Bilgin, MW Marcellin, J Serra-Sagrista & JA Storer (eds), Proceedings - DCC 2020: Data Compression Conference., 9105700, Data Compression Conference Proceedings, vol. 2020-March, Institute of Electrical and Electronics Engineers Inc., pp. 3-12, 2020 Data Compression Conference, DCC 2020, Snowbird, United States, 3/24/20. https://doi.org/10.1109/DCC47342.2020.00008

Diao E, Ding J, Tarokh V. DRASIC: Distributed recurrent autoencoder for scalable image compression. In Bilgin A, Marcellin MW, Serra-Sagrista J, Storer JA, editors, Proceedings - DCC 2020: Data Compression Conference. Institute of Electrical and Electronics Engineers Inc. 2020. p. 3-12. 9105700. (Data Compression Conference Proceedings). doi: 10.1109/DCC47342.2020.00008

@inproceedings{2474d25744dc4bf782ef48a7ea721166,

title = "DRASIC: Distributed recurrent autoencoder for scalable image compression",

abstract = "We propose a new architecture for distributed image compression from a group of distributed data sources. The work is motivated by practical needs of data-driven codec design, low power consumption, robustness, and data privacy. The proposed architecture, which we refer to as Distributed Recurrent Autoencoder for Scalable Image Compression (DRASIC), is able to train distributed encoders and one joint decoder on correlated data sources. Its compression capability is much better than the method of training codecs separately. Meanwhile, the performance of our distributed system with 10 distributed sources is only within 2 dB peak signal-to-noise ratio (PSNR) of the performance of a single codec trained with all data sources. We experiment distributed sources with different correlations and show how our data-driven methodology well matches the Slepian-Wolf Theorem in Distributed Source Coding (DSC). To the best of our knowledge, this is the first data-driven DSC framework for general distributed code design with deep learning.",

author = "Enmao Diao and Jie Ding and Vahid Tarokh",

note = "Funding Information: This work was supported in part by Office of Naval Research Grant No. N00014-18-1-2244. We provide our implementation at https://github.com/dem123456789/ Distributed-Recurrent-Autoencoder-for-Scalable-Image-Compression Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 Data Compression Conference, DCC 2020 ; Conference date: 24-03-2020 Through 27-03-2020",

year = "2020",

month = mar,

doi = "10.1109/DCC47342.2020.00008",

language = "English (US)",

series = "Data Compression Conference Proceedings",

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

pages = "3--12",

editor = "Ali Bilgin and Marcellin, {Michael W.} and Joan Serra-Sagrista and Storer, {James A.}",

booktitle = "Proceedings - DCC 2020",

}

TY - GEN

T1 - DRASIC

T2 - 2020 Data Compression Conference, DCC 2020

AU - Diao, Enmao

AU - Ding, Jie

AU - Tarokh, Vahid

N1 - Funding Information: This work was supported in part by Office of Naval Research Grant No. N00014-18-1-2244. We provide our implementation at https://github.com/dem123456789/ Distributed-Recurrent-Autoencoder-for-Scalable-Image-Compression Publisher Copyright: © 2020 IEEE.

PY - 2020/3

Y1 - 2020/3

N2 - We propose a new architecture for distributed image compression from a group of distributed data sources. The work is motivated by practical needs of data-driven codec design, low power consumption, robustness, and data privacy. The proposed architecture, which we refer to as Distributed Recurrent Autoencoder for Scalable Image Compression (DRASIC), is able to train distributed encoders and one joint decoder on correlated data sources. Its compression capability is much better than the method of training codecs separately. Meanwhile, the performance of our distributed system with 10 distributed sources is only within 2 dB peak signal-to-noise ratio (PSNR) of the performance of a single codec trained with all data sources. We experiment distributed sources with different correlations and show how our data-driven methodology well matches the Slepian-Wolf Theorem in Distributed Source Coding (DSC). To the best of our knowledge, this is the first data-driven DSC framework for general distributed code design with deep learning.

AB - We propose a new architecture for distributed image compression from a group of distributed data sources. The work is motivated by practical needs of data-driven codec design, low power consumption, robustness, and data privacy. The proposed architecture, which we refer to as Distributed Recurrent Autoencoder for Scalable Image Compression (DRASIC), is able to train distributed encoders and one joint decoder on correlated data sources. Its compression capability is much better than the method of training codecs separately. Meanwhile, the performance of our distributed system with 10 distributed sources is only within 2 dB peak signal-to-noise ratio (PSNR) of the performance of a single codec trained with all data sources. We experiment distributed sources with different correlations and show how our data-driven methodology well matches the Slepian-Wolf Theorem in Distributed Source Coding (DSC). To the best of our knowledge, this is the first data-driven DSC framework for general distributed code design with deep learning.

UR - http://www.scopus.com/inward/record.url?scp=85086869718&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85086869718&partnerID=8YFLogxK

U2 - 10.1109/DCC47342.2020.00008

DO - 10.1109/DCC47342.2020.00008

M3 - Conference contribution

AN - SCOPUS:85086869718

T3 - Data Compression Conference Proceedings

SP - 3

EP - 12

BT - Proceedings - DCC 2020

A2 - Bilgin, Ali

A2 - Marcellin, Michael W.

A2 - Serra-Sagrista, Joan

A2 - Storer, James A.

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 24 March 2020 through 27 March 2020

ER -

DRASIC: Distributed recurrent autoencoder for scalable image compression

Abstract

Publication series

Conference

Bibliographical note

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this