Enabling Image Recognition on Constrained Devices Using Neural Network Pruning and a CycleGAN
Lidfelt, August ; Isaksson, Daniel ; Hedlund, Ludwig ; Åberg, Simon ; Borg, Markus and Larsson, Erik LU
(2020)
First international workshop on Internet of Things for Emergency Management (IoT4Emergency)
- Abstract
- Smart cameras are increasingly used in surveillance solutions in public spaces. Contemporary computer vision applications can be used to recognize events that require intervention by emergency services. Smart cameras can be mounted in locations where citizens feel particularly unsafe, e.g., pathways and underpasses with a history of incidents. One promising approach for smart cameras is edge AI, i.e., deploying AI technology on IoT devices. However, implementing resource-demanding technology such as image recognition using deep neural networks (DNN) on constrained devices is a substantial challenge. In this paper, we explore two approaches to reduce the need for compute in contemporary image recognition in an underpass. First, we showcase... (More)
- Smart cameras are increasingly used in surveillance solutions in public spaces. Contemporary computer vision applications can be used to recognize events that require intervention by emergency services. Smart cameras can be mounted in locations where citizens feel particularly unsafe, e.g., pathways and underpasses with a history of incidents. One promising approach for smart cameras is edge AI, i.e., deploying AI technology on IoT devices. However, implementing resource-demanding technology such as image recognition using deep neural networks (DNN) on constrained devices is a substantial challenge. In this paper, we explore two approaches to reduce the need for compute in contemporary image recognition in an underpass. First, we showcase successful neural network pruning, i.e., we retain comparable classification accuracy with only 1.1% of the neurons remaining from the state-of-the-art DNN architecture. Second, we demonstrate how a CycleGAN can be used to transform out-of-distribution images to the operational design domain. We posit that both pruning and CycleGANs are promising enablers for efficient edge AI in smart cameras. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/f64186b8-d10f-464f-9bff-50ee484d0349
- author
- Lidfelt, August
; Isaksson, Daniel
; Hedlund, Ludwig
; Åberg, Simon
; Borg, Markus
and Larsson, Erik
LU
- organization
- publishing date
- 2020
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- smart camera, image recognition, neural network pruning, genera- tive adversarial network, edge AI
- host publication
- 1st international workshop on Internet of Things for Emergency Management
- pages
- 13 pages
- publisher
- Association for Computing Machinery (ACM)
- conference name
- First international workshop on Internet of Things for Emergency Management (IoT4Emergency)
- conference location
- Malmö, Sweden
- conference dates
- 2020-10-06 - 2020-10-06
- external identifiers
-
- scopus:85117541633
- ISBN
- 9781450388207
- DOI
- 10.1145/3423423.3423437
- project
- AIQ Meta-Testbed
- language
- English
- LU publication?
- yes
- id
- f64186b8-d10f-464f-9bff-50ee484d0349
- date added to LUP
- 2020-09-16 15:37:47
- date last changed
- 2025-04-04 13:55:18
@inproceedings{f64186b8-d10f-464f-9bff-50ee484d0349,
abstract = {{Smart cameras are increasingly used in surveillance solutions in public spaces. Contemporary computer vision applications can be used to recognize events that require intervention by emergency services. Smart cameras can be mounted in locations where citizens feel particularly unsafe, e.g., pathways and underpasses with a history of incidents. One promising approach for smart cameras is edge AI, i.e., deploying AI technology on IoT devices. However, implementing resource-demanding technology such as image recognition using deep neural networks (DNN) on constrained devices is a substantial challenge. In this paper, we explore two approaches to reduce the need for compute in contemporary image recognition in an underpass. First, we showcase successful neural network pruning, i.e., we retain comparable classification accuracy with only 1.1% of the neurons remaining from the state-of-the-art DNN architecture. Second, we demonstrate how a CycleGAN can be used to transform out-of-distribution images to the operational design domain. We posit that both pruning and CycleGANs are promising enablers for efficient edge AI in smart cameras.}},
author = {{Lidfelt, August and Isaksson, Daniel and Hedlund, Ludwig and Åberg, Simon and Borg, Markus and Larsson, Erik}},
booktitle = {{1st international workshop on Internet of Things for Emergency Management}},
isbn = {{9781450388207}},
keywords = {{smart camera, image recognition, neural network pruning, genera- tive adversarial network, edge AI}},
language = {{eng}},
publisher = {{Association for Computing Machinery (ACM)}},
title = {{Enabling Image Recognition on Constrained Devices Using Neural Network Pruning and a CycleGAN}},
url = {{https://lup.lub.lu.se/search/files/83812625/2020_IoT4Emergency_Lidfeldt_Enabling.pdf}},
doi = {{10.1145/3423423.3423437}},
year = {{2020}},
}