MRIM: Enabling Mixed-Resolution Imaging for low-power pervasive vision tasks

While many pervasive computing applications increasingly utilize real-time context extracted from a vision sensing infrastructure, the high energy overhead of DNN-based vision sensing pipelines remains a challenge for sustainable in-the-wild deployment. One common approach to reducing such energy ov...

Full description

Saved in:
Bibliographic Details
Main Authors: WU, Jiyan, SUBASHARAN, Vithurson, TRAN, Tuan, MISRA, Archan
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2022
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/7165
https://ink.library.smu.edu.sg/context/sis_research/article/8167/viewcontent/a5_wu_final_MRIM.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-8167
record_format dspace
spelling sg-smu-ink.sis_research-81672023-08-04T05:33:54Z MRIM: Enabling Mixed-Resolution Imaging for low-power pervasive vision tasks WU, Jiyan SUBASHARAN, Vithurson TRAN, Tuan MISRA, Archan While many pervasive computing applications increasingly utilize real-time context extracted from a vision sensing infrastructure, the high energy overhead of DNN-based vision sensing pipelines remains a challenge for sustainable in-the-wild deployment. One common approach to reducing such energy overheads is the capture and transmission of lower-resolution images to an edge node (where the DNN inferencing task is executed), but this results in an accuracy-vs-energy tradeoff, as the DNN inference accuracy typically degrades with a drop in resolution. In this work, we introduce MRIM, a simple but effective framework to tackle this tradeoff. Under MRIM, the vision sensor platform first executes a lightweight preprocessing step to determine the saliency of different sub-regions within a single captured image frame, and then performs a saliency-aware non-uniform downscaling of individual sub-regions to produce a “mixed-resolution” image. We describe two novel low-complexity algorithms that the sensor platform can use to quickly compute suitable resolution choices for different regions under different energy/accuracy constraints. Experimental studies, involving object detection tasks evaluated traces from two benchmark urban monitoring datasets as well as a prototype Raspberry Pi-based MRIM implementation, demonstrate MRIM’s efficacy: even with unoptimized embedded platform, MRIM can provide system energy savings of 35+% or increase task accuracy by 8+%, over conventional baselines of uniform resolution downscaling or image encoding, while supporting high throughput. 2022-03-01T08:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/7165 info:doi/10.1109/PerCom53586.2022.9762398 https://ink.library.smu.edu.sg/context/sis_research/article/8167/viewcontent/a5_wu_final_MRIM.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 Mixed resolution imaging tasks energy consumption Graphics and Human Computer Interfaces Software Engineering
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Mixed resolution
imaging tasks
energy consumption
Graphics and Human Computer Interfaces
Software Engineering
spellingShingle Mixed resolution
imaging tasks
energy consumption
Graphics and Human Computer Interfaces
Software Engineering
WU, Jiyan
SUBASHARAN, Vithurson
TRAN, Tuan
MISRA, Archan
MRIM: Enabling Mixed-Resolution Imaging for low-power pervasive vision tasks
description While many pervasive computing applications increasingly utilize real-time context extracted from a vision sensing infrastructure, the high energy overhead of DNN-based vision sensing pipelines remains a challenge for sustainable in-the-wild deployment. One common approach to reducing such energy overheads is the capture and transmission of lower-resolution images to an edge node (where the DNN inferencing task is executed), but this results in an accuracy-vs-energy tradeoff, as the DNN inference accuracy typically degrades with a drop in resolution. In this work, we introduce MRIM, a simple but effective framework to tackle this tradeoff. Under MRIM, the vision sensor platform first executes a lightweight preprocessing step to determine the saliency of different sub-regions within a single captured image frame, and then performs a saliency-aware non-uniform downscaling of individual sub-regions to produce a “mixed-resolution” image. We describe two novel low-complexity algorithms that the sensor platform can use to quickly compute suitable resolution choices for different regions under different energy/accuracy constraints. Experimental studies, involving object detection tasks evaluated traces from two benchmark urban monitoring datasets as well as a prototype Raspberry Pi-based MRIM implementation, demonstrate MRIM’s efficacy: even with unoptimized embedded platform, MRIM can provide system energy savings of 35+% or increase task accuracy by 8+%, over conventional baselines of uniform resolution downscaling or image encoding, while supporting high throughput.
format text
author WU, Jiyan
SUBASHARAN, Vithurson
TRAN, Tuan
MISRA, Archan
author_facet WU, Jiyan
SUBASHARAN, Vithurson
TRAN, Tuan
MISRA, Archan
author_sort WU, Jiyan
title MRIM: Enabling Mixed-Resolution Imaging for low-power pervasive vision tasks
title_short MRIM: Enabling Mixed-Resolution Imaging for low-power pervasive vision tasks
title_full MRIM: Enabling Mixed-Resolution Imaging for low-power pervasive vision tasks
title_fullStr MRIM: Enabling Mixed-Resolution Imaging for low-power pervasive vision tasks
title_full_unstemmed MRIM: Enabling Mixed-Resolution Imaging for low-power pervasive vision tasks
title_sort mrim: enabling mixed-resolution imaging for low-power pervasive vision tasks
publisher Institutional Knowledge at Singapore Management University
publishDate 2022
url https://ink.library.smu.edu.sg/sis_research/7165
https://ink.library.smu.edu.sg/context/sis_research/article/8167/viewcontent/a5_wu_final_MRIM.pdf
_version_ 1773551431616299008