Skip to main navigation Skip to search Skip to main content

Adder-Only Convolutional Neural Network with Binary Input Image

  • Indian Institute of Technology Kharagpur
  • The University of Tokyo

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Convolutional neural networks (CNNs) have performed exceptionally well on a variety of image classification tasks but need significant amount of memory and computational resources. In this paper, we propose an adder-only CNN (AO-CNN) inference engine, that has its weights and/or outputs at each layer reduced to either powers of two or zero, thus replacing the multiplication operations with a simple right or left shift and hence reducing the computation significantly. The proposed AO-CNN architecture, demonstrated using three sets of data-sets, viz., MNIST, EMNIST, and SVHN, performs with ≈ 80% accuracy or more, while using minimal hardware and taking only binary input images, i.e., 1 and 0. This could pave the way for realization of image-sensors with 1-bit resolution images and the corresponding CNN based classification engine being integrated into low-power internet-of-things (IoT) devices.

Original languageEnglish
Title of host publication2019 IEEE 62nd International Midwest Symposium on Circuits and Systems, MWSCAS 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages319-322
Number of pages4
ISBN (Electronic)9781728127880
DOIs
StatePublished - Aug 2019
Event62nd IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2019 - Dallas, United States
Duration: Aug 4 2019Aug 7 2019

Publication series

NameMidwest Symposium on Circuits and Systems
Volume2019-August
ISSN (Print)1548-3746

Conference

Conference62nd IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2019
Country/TerritoryUnited States
CityDallas
Period08/4/1908/7/19

Fingerprint

Dive into the research topics of 'Adder-Only Convolutional Neural Network with Binary Input Image'. Together they form a unique fingerprint.

Cite this