On-chip photonic convolutional accelerator for image processing

Junwei Cheng; Yuhe Zhao; Yanxian Wei; Wenkai Zhang; Hailong Zhou; Dongmei Huang; Feng Li; Alex Wai; Jianji Dong; Xinliang Zhang

doi:10.1364/OECC.2021.W4C.6

On-chip photonic convolutional accelerator for image processing

Junwei Cheng, Yuhe Zhao, Yanxian Wei, Wenkai Zhang, Hailong Zhou, Dongmei Huang, Feng Li, Alex Wai, Jianji Dong^*, Xinliang Zhang

^*Corresponding author for this work

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

2 Citations (Scopus)

Abstract

We demonstrate an on-chip photonic convolutional accelerator, which can perform configurable optical convolution operations to achieve specific image processing. It indicates great potential for on-chip image processing in AI hardware acceleration.

Original language	English
Title of host publication	Proceedings of 2021 Opto-Electronics and Communications Conference
Publisher	IEEE
ISBN (Electronic)	9781943580927
DOIs	https://doi.org/10.1364/OECC.2021.W4C.6
Publication status	Published - Jul 2021
Event	2021 Opto-Electronics and Communications Conference, OECC 2021 - Online, Hong Kong Duration: 3 Jul 2021 → 7 Jul 2021 http://www.oecc2021.org/ (Link to conference website) http://www.oecc2021.org/files/OECC2021%20FullProgram%200710v6.pdf (Link to conference programme)

Publication series

Name	Opto-Electronics and Communications Conference, OECC

Conference

Conference	2021 Opto-Electronics and Communications Conference, OECC 2021
Country/Territory	Hong Kong
Period	3/07/21 → 7/07/21
Internet address	http://www.oecc2021.org/ (Link to conference website) http://www.oecc2021.org/files/OECC2021%20FullProgram%200710v6.pdf (Link to conference programme)

Scopus Subject Areas

Computer Networks and Communications
Electrical and Electronic Engineering
Electronic, Optical and Magnetic Materials
Atomic and Molecular Physics, and Optics
Artificial Intelligence

Access to Document

10.1364/OECC.2021.W4C.6

Cite this

@inproceedings{904c1bdb32be4b23a75cf62b859fb27b,

title = "On-chip photonic convolutional accelerator for image processing",

abstract = "We demonstrate an on-chip photonic convolutional accelerator, which can perform configurable optical convolution operations to achieve specific image processing. It indicates great potential for on-chip image processing in AI hardware acceleration.",

author = "Junwei Cheng and Yuhe Zhao and Yanxian Wei and Wenkai Zhang and Hailong Zhou and Dongmei Huang and Feng Li and Alex Wai and Jianji Dong and Xinliang Zhang",

note = "Funding Information: This work was partially supported by National Key R&D Program of China (2019YFB1803900), the National Natural Science Foundation of China (61805090, 62075075), the China Postdoctoral Science Foundation (2017M622419) and Research Grants Council, University Grants Committee of Hong Kong SAR (PolyU152241/18E). Publisher Copyright: {\textcopyright} 2021 The Author(s) ; 2021 Opto-Electronics and Communications Conference, OECC 2021 ; Conference date: 03-07-2021 Through 07-07-2021",

year = "2021",

month = jul,

doi = "10.1364/OECC.2021.W4C.6",

language = "English",

series = "Opto-Electronics and Communications Conference, OECC ",

publisher = "IEEE",

booktitle = "Proceedings of 2021 Opto-Electronics and Communications Conference",

address = "United States",

url = "http://www.oecc2021.org/, http://www.oecc2021.org/files/OECC2021%20FullProgram%200710v6.pdf",

}

Cheng, J, Zhao, Y, Wei, Y, Zhang, W, Zhou, H, Huang, D, Li, F, Wai, A, Dong, J & Zhang, X 2021, On-chip photonic convolutional accelerator for image processing. in Proceedings of 2021 Opto-Electronics and Communications Conference., W4C.6, Opto-Electronics and Communications Conference, OECC , IEEE, 2021 Opto-Electronics and Communications Conference, OECC 2021, Hong Kong, 3/07/21. https://doi.org/10.1364/OECC.2021.W4C.6

TY - GEN

T1 - On-chip photonic convolutional accelerator for image processing

AU - Cheng, Junwei

AU - Zhao, Yuhe

AU - Wei, Yanxian

AU - Zhang, Wenkai

AU - Zhou, Hailong

AU - Huang, Dongmei

AU - Li, Feng

AU - Wai, Alex

AU - Dong, Jianji

AU - Zhang, Xinliang

N1 - Funding Information: This work was partially supported by National Key R&D Program of China (2019YFB1803900), the National Natural Science Foundation of China (61805090, 62075075), the China Postdoctoral Science Foundation (2017M622419) and Research Grants Council, University Grants Committee of Hong Kong SAR (PolyU152241/18E). Publisher Copyright: © 2021 The Author(s)

PY - 2021/7

Y1 - 2021/7

N2 - We demonstrate an on-chip photonic convolutional accelerator, which can perform configurable optical convolution operations to achieve specific image processing. It indicates great potential for on-chip image processing in AI hardware acceleration.

AB - We demonstrate an on-chip photonic convolutional accelerator, which can perform configurable optical convolution operations to achieve specific image processing. It indicates great potential for on-chip image processing in AI hardware acceleration.

UR - http://www.scopus.com/inward/record.url?scp=85139011015&partnerID=8YFLogxK

U2 - 10.1364/OECC.2021.W4C.6

DO - 10.1364/OECC.2021.W4C.6

M3 - Conference proceeding

AN - SCOPUS:85119595864

T3 - Opto-Electronics and Communications Conference, OECC

BT - Proceedings of 2021 Opto-Electronics and Communications Conference

PB - IEEE

T2 - 2021 Opto-Electronics and Communications Conference, OECC 2021

Y2 - 3 July 2021 through 7 July 2021

ER -

On-chip photonic convolutional accelerator for image processing

Abstract

Publication series

Conference

Scopus Subject Areas

Access to Document

Other files and links

Fingerprint

Cite this