DEMO: DeepMon - Building mobile GPU Deep learning models for continuous vision applications

Deep learning has revolutionized vision sensing applications in terms of accuracy comparing to other techniques. Its breakthrough comes from the ability to extract complex high level features directly from sensor data. However, deep learning models are still yet to be natively supported on mobile de...

Full description

Saved in:
Bibliographic Details
Main Authors: HUYNH, Loc Nguyen, BALAN, Rajesh Krishna, LEE, Youngki
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2017
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/3672
https://ink.library.smu.edu.sg/context/sis_research/article/4674/viewcontent/p186_huynh.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-4674
record_format dspace
spelling sg-smu-ink.sis_research-46742018-12-19T06:45:14Z DEMO: DeepMon - Building mobile GPU Deep learning models for continuous vision applications HUYNH, Loc Nguyen BALAN, Rajesh Krishna LEE, Youngki Deep learning has revolutionized vision sensing applications in terms of accuracy comparing to other techniques. Its breakthrough comes from the ability to extract complex high level features directly from sensor data. However, deep learning models are still yet to be natively supported on mobile devices due to high computational requirements. In this paper, we present DeepMon, a next generation of DeepSense [1] framework, to enable deep learning models on conventional mobile devices (e.g. Samsung Galaxy S7) for continuous vision sensing applications. Firstly, Deep-Mon exploits similarity between consecutive video frames for intermediate data caching within models to enhance inference latency. Secondly, DeepMon leverages approximation technique (e.g. Tucker decomposition) to build up approximated models with negligible impact on accuracy. Thirdly, DeepMon ofloads heavy computation onto integrated mobile GPU to significantly reduce execution time of the model. 2017-06-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/3672 info:doi/10.1145/3081333.3089331 https://ink.library.smu.edu.sg/context/sis_research/article/4674/viewcontent/p186_huynh.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Continuous vision Deep learning Mobile GPU Mobile sensing Hardware Systems Software Engineering
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Continuous vision
Deep learning
Mobile GPU
Mobile sensing
Hardware Systems
Software Engineering
spellingShingle Continuous vision
Deep learning
Mobile GPU
Mobile sensing
Hardware Systems
Software Engineering
HUYNH, Loc Nguyen
BALAN, Rajesh Krishna
LEE, Youngki
DEMO: DeepMon - Building mobile GPU Deep learning models for continuous vision applications
description Deep learning has revolutionized vision sensing applications in terms of accuracy comparing to other techniques. Its breakthrough comes from the ability to extract complex high level features directly from sensor data. However, deep learning models are still yet to be natively supported on mobile devices due to high computational requirements. In this paper, we present DeepMon, a next generation of DeepSense [1] framework, to enable deep learning models on conventional mobile devices (e.g. Samsung Galaxy S7) for continuous vision sensing applications. Firstly, Deep-Mon exploits similarity between consecutive video frames for intermediate data caching within models to enhance inference latency. Secondly, DeepMon leverages approximation technique (e.g. Tucker decomposition) to build up approximated models with negligible impact on accuracy. Thirdly, DeepMon ofloads heavy computation onto integrated mobile GPU to significantly reduce execution time of the model.
format text
author HUYNH, Loc Nguyen
BALAN, Rajesh Krishna
LEE, Youngki
author_facet HUYNH, Loc Nguyen
BALAN, Rajesh Krishna
LEE, Youngki
author_sort HUYNH, Loc Nguyen
title DEMO: DeepMon - Building mobile GPU Deep learning models for continuous vision applications
title_short DEMO: DeepMon - Building mobile GPU Deep learning models for continuous vision applications
title_full DEMO: DeepMon - Building mobile GPU Deep learning models for continuous vision applications
title_fullStr DEMO: DeepMon - Building mobile GPU Deep learning models for continuous vision applications
title_full_unstemmed DEMO: DeepMon - Building mobile GPU Deep learning models for continuous vision applications
title_sort demo: deepmon - building mobile gpu deep learning models for continuous vision applications
publisher Institutional Knowledge at Singapore Management University
publishDate 2017
url https://ink.library.smu.edu.sg/sis_research/3672
https://ink.library.smu.edu.sg/context/sis_research/article/4674/viewcontent/p186_huynh.pdf
_version_ 1770573636183261184