CondenseNeXtV2: Light-Weight Modern Image Classifier Utilizing Self-Querying Augmentation Policies
Artificial Intelligence (AI) combines computer science and robust datasets to mimic natural intelligence demonstrated by human beings to aid in problem-solving and decision-making involving consciousness up to a certain extent. From Apple’s virtual personal assistant, Siri, to Tesla’s self-driving c...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-02-01
|
Series: | Journal of Low Power Electronics and Applications |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9268/12/1/8 |
_version_ | 1797470384321200128 |
---|---|
author | Priyank Kalgaonkar Mohamed El-Sharkawy |
author_facet | Priyank Kalgaonkar Mohamed El-Sharkawy |
author_sort | Priyank Kalgaonkar |
collection | DOAJ |
description | Artificial Intelligence (AI) combines computer science and robust datasets to mimic natural intelligence demonstrated by human beings to aid in problem-solving and decision-making involving consciousness up to a certain extent. From Apple’s virtual personal assistant, Siri, to Tesla’s self-driving cars, research and development in the field of AI is progressing rapidly along with privacy concerns surrounding the usage and storage of user data on external servers which has further fueled the need of modern ultra-efficient AI networks and algorithms. The scope of the work presented within this paper focuses on introducing a modern image classifier which is a light-weight and ultra-efficient CNN intended to be deployed on local embedded systems, also known as edge devices, for general-purpose usage. This work is an extension of the award-winning paper entitled ‘CondenseNeXt: An Ultra-Efficient Deep Neural Network for Embedded Systems’ published for the 2021 IEEE 11th Annual Computing and Communication Workshop and Conference (CCWC). The proposed neural network dubbed CondenseNeXtV2 utilizes a new self-querying augmentation policy technique on the target dataset along with adaption to the latest version of PyTorch framework and activation functions resulting in improved efficiency in image classification computation and accuracy. Finally, we deploy the trained weights of CondenseNeXtV2 on NXP BlueBox which is an edge device designed to serve as a development platform for self-driving cars, and conclusions will be extrapolated accordingly. |
first_indexed | 2024-03-09T19:35:48Z |
format | Article |
id | doaj.art-a5fe436fcd384344abfe48849f0def8b |
institution | Directory Open Access Journal |
issn | 2079-9268 |
language | English |
last_indexed | 2024-03-09T19:35:48Z |
publishDate | 2022-02-01 |
publisher | MDPI AG |
record_format | Article |
series | Journal of Low Power Electronics and Applications |
spelling | doaj.art-a5fe436fcd384344abfe48849f0def8b2023-11-24T01:55:53ZengMDPI AGJournal of Low Power Electronics and Applications2079-92682022-02-01121810.3390/jlpea12010008CondenseNeXtV2: Light-Weight Modern Image Classifier Utilizing Self-Querying Augmentation PoliciesPriyank Kalgaonkar0Mohamed El-Sharkawy1Department of Electrical and Computer Engineering, Purdue School of Engineering and Technology, Indianapolis, IN 46254, USADepartment of Electrical and Computer Engineering, Purdue School of Engineering and Technology, Indianapolis, IN 46254, USAArtificial Intelligence (AI) combines computer science and robust datasets to mimic natural intelligence demonstrated by human beings to aid in problem-solving and decision-making involving consciousness up to a certain extent. From Apple’s virtual personal assistant, Siri, to Tesla’s self-driving cars, research and development in the field of AI is progressing rapidly along with privacy concerns surrounding the usage and storage of user data on external servers which has further fueled the need of modern ultra-efficient AI networks and algorithms. The scope of the work presented within this paper focuses on introducing a modern image classifier which is a light-weight and ultra-efficient CNN intended to be deployed on local embedded systems, also known as edge devices, for general-purpose usage. This work is an extension of the award-winning paper entitled ‘CondenseNeXt: An Ultra-Efficient Deep Neural Network for Embedded Systems’ published for the 2021 IEEE 11th Annual Computing and Communication Workshop and Conference (CCWC). The proposed neural network dubbed CondenseNeXtV2 utilizes a new self-querying augmentation policy technique on the target dataset along with adaption to the latest version of PyTorch framework and activation functions resulting in improved efficiency in image classification computation and accuracy. Finally, we deploy the trained weights of CondenseNeXtV2 on NXP BlueBox which is an edge device designed to serve as a development platform for self-driving cars, and conclusions will be extrapolated accordingly.https://www.mdpi.com/2079-9268/12/1/8CondenseNeXtconvolutional neural networkcomputer visionembedded systemsedge devicesimage classification |
spellingShingle | Priyank Kalgaonkar Mohamed El-Sharkawy CondenseNeXtV2: Light-Weight Modern Image Classifier Utilizing Self-Querying Augmentation Policies Journal of Low Power Electronics and Applications CondenseNeXt convolutional neural network computer vision embedded systems edge devices image classification |
title | CondenseNeXtV2: Light-Weight Modern Image Classifier Utilizing Self-Querying Augmentation Policies |
title_full | CondenseNeXtV2: Light-Weight Modern Image Classifier Utilizing Self-Querying Augmentation Policies |
title_fullStr | CondenseNeXtV2: Light-Weight Modern Image Classifier Utilizing Self-Querying Augmentation Policies |
title_full_unstemmed | CondenseNeXtV2: Light-Weight Modern Image Classifier Utilizing Self-Querying Augmentation Policies |
title_short | CondenseNeXtV2: Light-Weight Modern Image Classifier Utilizing Self-Querying Augmentation Policies |
title_sort | condensenextv2 light weight modern image classifier utilizing self querying augmentation policies |
topic | CondenseNeXt convolutional neural network computer vision embedded systems edge devices image classification |
url | https://www.mdpi.com/2079-9268/12/1/8 |
work_keys_str_mv | AT priyankkalgaonkar condensenextv2lightweightmodernimageclassifierutilizingselfqueryingaugmentationpolicies AT mohamedelsharkawy condensenextv2lightweightmodernimageclassifierutilizingselfqueryingaugmentationpolicies |