CrowdVision: A Computing Platform for Video Crowdprocessing Using Deep Learning

Zongqing Lu, Kevin Chan, Shiliang Pu, Thomas F. La Porta

Research output: Contribution to journalArticle

Abstract

Mobile devices such as smartphones are enabling users to generate and share videos with increasing rates. In some cases, these videos may contain valuable information, which can be exploited for a variety of purposes. However, instead of centrally collecting and processing videos for information retrieval, we consider crowdprocessing videos, where each mobile device locally processes stored videos. While the computational capability of mobile devices continues to improve, processing videos using deep learning, i.e., convolutional neural networks, is still a demanding task for mobile devices. To this end, we design and build CrowdVision, a computing platform that enables mobile devices to crowdprocess videos using deep learning in a distributed and energy-efficient manner leveraging cloud offload. CrowdVision can quickly and efficiently process videos with offload under various settings and different network connections and greatly outperform the existing computation offload framework (e.g., with a 2× speed-up). In doing so, CrowdVision tackles several challenges: (i) how to exploit the characteristics of the computing of deep learning for video processing; (ii) how to parallelize processing and offloading for acceleration; and (iii) how to optimize both time and energy at runtime by just determining the right moments to offload.

Original languageEnglish (US)
Article number8428426
Pages (from-to)1513-1526
Number of pages14
JournalIEEE Transactions on Mobile Computing
Volume18
Issue number7
DOIs
StatePublished - Jul 1 2019

Fingerprint

Mobile devices
Processing
Smartphones
Information retrieval
Deep learning
Neural networks

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Cite this

@article{4453186fc986484a81eb65479353d53f,
title = "CrowdVision: A Computing Platform for Video Crowdprocessing Using Deep Learning",
abstract = "Mobile devices such as smartphones are enabling users to generate and share videos with increasing rates. In some cases, these videos may contain valuable information, which can be exploited for a variety of purposes. However, instead of centrally collecting and processing videos for information retrieval, we consider crowdprocessing videos, where each mobile device locally processes stored videos. While the computational capability of mobile devices continues to improve, processing videos using deep learning, i.e., convolutional neural networks, is still a demanding task for mobile devices. To this end, we design and build CrowdVision, a computing platform that enables mobile devices to crowdprocess videos using deep learning in a distributed and energy-efficient manner leveraging cloud offload. CrowdVision can quickly and efficiently process videos with offload under various settings and different network connections and greatly outperform the existing computation offload framework (e.g., with a 2× speed-up). In doing so, CrowdVision tackles several challenges: (i) how to exploit the characteristics of the computing of deep learning for video processing; (ii) how to parallelize processing and offloading for acceleration; and (iii) how to optimize both time and energy at runtime by just determining the right moments to offload.",
author = "Zongqing Lu and Kevin Chan and Shiliang Pu and {La Porta}, {Thomas F.}",
year = "2019",
month = "7",
day = "1",
doi = "10.1109/TMC.2018.2864212",
language = "English (US)",
volume = "18",
pages = "1513--1526",
journal = "IEEE Transactions on Mobile Computing",
issn = "1536-1233",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "7",

}

CrowdVision : A Computing Platform for Video Crowdprocessing Using Deep Learning. / Lu, Zongqing; Chan, Kevin; Pu, Shiliang; La Porta, Thomas F.

In: IEEE Transactions on Mobile Computing, Vol. 18, No. 7, 8428426, 01.07.2019, p. 1513-1526.

Research output: Contribution to journalArticle

TY - JOUR

T1 - CrowdVision

T2 - A Computing Platform for Video Crowdprocessing Using Deep Learning

AU - Lu, Zongqing

AU - Chan, Kevin

AU - Pu, Shiliang

AU - La Porta, Thomas F.

PY - 2019/7/1

Y1 - 2019/7/1

N2 - Mobile devices such as smartphones are enabling users to generate and share videos with increasing rates. In some cases, these videos may contain valuable information, which can be exploited for a variety of purposes. However, instead of centrally collecting and processing videos for information retrieval, we consider crowdprocessing videos, where each mobile device locally processes stored videos. While the computational capability of mobile devices continues to improve, processing videos using deep learning, i.e., convolutional neural networks, is still a demanding task for mobile devices. To this end, we design and build CrowdVision, a computing platform that enables mobile devices to crowdprocess videos using deep learning in a distributed and energy-efficient manner leveraging cloud offload. CrowdVision can quickly and efficiently process videos with offload under various settings and different network connections and greatly outperform the existing computation offload framework (e.g., with a 2× speed-up). In doing so, CrowdVision tackles several challenges: (i) how to exploit the characteristics of the computing of deep learning for video processing; (ii) how to parallelize processing and offloading for acceleration; and (iii) how to optimize both time and energy at runtime by just determining the right moments to offload.

AB - Mobile devices such as smartphones are enabling users to generate and share videos with increasing rates. In some cases, these videos may contain valuable information, which can be exploited for a variety of purposes. However, instead of centrally collecting and processing videos for information retrieval, we consider crowdprocessing videos, where each mobile device locally processes stored videos. While the computational capability of mobile devices continues to improve, processing videos using deep learning, i.e., convolutional neural networks, is still a demanding task for mobile devices. To this end, we design and build CrowdVision, a computing platform that enables mobile devices to crowdprocess videos using deep learning in a distributed and energy-efficient manner leveraging cloud offload. CrowdVision can quickly and efficiently process videos with offload under various settings and different network connections and greatly outperform the existing computation offload framework (e.g., with a 2× speed-up). In doing so, CrowdVision tackles several challenges: (i) how to exploit the characteristics of the computing of deep learning for video processing; (ii) how to parallelize processing and offloading for acceleration; and (iii) how to optimize both time and energy at runtime by just determining the right moments to offload.

UR - http://www.scopus.com/inward/record.url?scp=85051390358&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051390358&partnerID=8YFLogxK

U2 - 10.1109/TMC.2018.2864212

DO - 10.1109/TMC.2018.2864212

M3 - Article

AN - SCOPUS:85051390358

VL - 18

SP - 1513

EP - 1526

JO - IEEE Transactions on Mobile Computing

JF - IEEE Transactions on Mobile Computing

SN - 1536-1233

IS - 7

M1 - 8428426

ER -