Benchmarking the Accuracy of Algorithms for Memory-Constrained Image Classification

Sebastian Müksch, Theo Olausson, John Wilhelm, Pavlos Andreadis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Convolutional Neural Networks, or CNNs, are the state of the art for image classification, but typically come at the cost of a large memory footprint. This limits their usefulness in edge computing applications, where memory is often a scarce resource. Recently, there has been significant progress in the field of image classification on such memory-constrained devices, with novel contributions like the ProtoNN, Bonsai and FastGRNN algorithms. These have been shown to reach up to 98.2% accuracy on optical character recognition using MNIST-10, with a memory footprint as little as 6KB. However, their potential on more complex multi-class and multi-channel image classification has yet to be determined. In this paper, we compare CNNs with ProtoNN, Bonsai and FastGRNN when applied to 3-channel image classification using CIFAR-10. For our analysis, we use the existing Direct Convolution algorithm to implement the CNNs memory-optimally and propose new methods of adjusting the FastGRNN model to work with multi-channel images. We extend the evaluation of each algorithm to a memory size budget of 8KB, 16KB, 32KB, 64KB and 128KB to show quantitatively that Direct Convolution CNNs perform best for all chosen budgets, with a top performance of 65.7% accuracy at a memory footprint of 58.23KB.
Original languageEnglish
Title of host publication2020 IEEE/ACM Symposium on Edge Computing (SEC)
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages400-404
Number of pages5
ISBN (Electronic)978-1-7281-5943-0
ISBN (Print)978-1-7281-5944-7
DOIs
Publication statusPublished - 22 Feb 2021
EventFifth ACM/IEEE Symposium on Edge Computing - Virtual, San Jose, United States
Duration: 11 Nov 202013 Nov 2020
http://acm-ieee-sec.org/2020/

Conference

ConferenceFifth ACM/IEEE Symposium on Edge Computing
Abbreviated titleSEC 2020
CountryUnited States
CitySan Jose
Period11/11/2013/11/20
Internet address

Fingerprint

Dive into the research topics of 'Benchmarking the Accuracy of Algorithms for Memory-Constrained Image Classification'. Together they form a unique fingerprint.

Cite this