Poster: Retrofitting computer vision libraries for concurrent support on mobile devices

Robert LiKamWa; Eddie Reyes; Lin Zhong

doi:10.1145/2639108.2642891

Poster: Retrofitting computer vision libraries for concurrent support on mobile devices

Robert LiKamWa, Eddie Reyes, Lin Zhong

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

While computer vision algorithms and libraries have enabled and accelerated the adoption of vision processing into mobile and wearable applications, vision is a resource-hungry operation, and is thus not efficient enough to run on multiple applications simultaneously. However, we observe that many vision algorithms share identical sets of frames and features to perform their analyses, computed from the same library calls. Leveraging this observation, we design a split-process architecture to retrofit existing vision libraries to allow applications to transparently share the computational, memory, and energy overhead of vision processing.

Original language	English (US)
Title of host publication	MobiCom 2014 - Proceedings of the 20th Annual
Publisher	Association for Computing Machinery
Pages	387-389
Number of pages	3
ISBN (Electronic)	9781450327831
DOIs	https://doi.org/10.1145/2639108.2642891
State	Published - Sep 7 2014
Externally published	Yes
Event	20th ACM Annual International Conference on Mobile Computing and Networking, MobiCom 2014 - Maui, United States Duration: Sep 7 2014 → Sep 11 2014

Publication series

Name	Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM

Other

Other	20th ACM Annual International Conference on Mobile Computing and Networking, MobiCom 2014
Country/Territory	United States
City	Maui
Period	9/7/14 → 9/11/14

Keywords

Remote execution
Shared memory management
Split-process

ASJC Scopus subject areas

Software
Hardware and Architecture
Computer Networks and Communications

Access to Document

10.1145/2639108.2642891

Cite this

LiKamWa, R., Reyes, E., & Zhong, L. (2014). Poster: Retrofitting computer vision libraries for concurrent support on mobile devices. In MobiCom 2014 - Proceedings of the 20th Annual (pp. 387-389). (Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM). Association for Computing Machinery. https://doi.org/10.1145/2639108.2642891

Poster: Retrofitting computer vision libraries for concurrent support on mobile devices. / LiKamWa, Robert; Reyes, Eddie; Zhong, Lin.
MobiCom 2014 - Proceedings of the 20th Annual. Association for Computing Machinery, 2014. p. 387-389 (Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

LiKamWa, R, Reyes, E & Zhong, L 2014, Poster: Retrofitting computer vision libraries for concurrent support on mobile devices. in MobiCom 2014 - Proceedings of the 20th Annual. Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM, Association for Computing Machinery, pp. 387-389, 20th ACM Annual International Conference on Mobile Computing and Networking, MobiCom 2014, Maui, United States, 9/7/14. https://doi.org/10.1145/2639108.2642891

@inproceedings{b93a08c737c548bead90135cc74d8554,

title = "Poster: Retrofitting computer vision libraries for concurrent support on mobile devices",

abstract = "While computer vision algorithms and libraries have enabled and accelerated the adoption of vision processing into mobile and wearable applications, vision is a resource-hungry operation, and is thus not efficient enough to run on multiple applications simultaneously. However, we observe that many vision algorithms share identical sets of frames and features to perform their analyses, computed from the same library calls. Leveraging this observation, we design a split-process architecture to retrofit existing vision libraries to allow applications to transparently share the computational, memory, and energy overhead of vision processing.",

keywords = "Remote execution, Shared memory management, Split-process",

author = "Robert LiKamWa and Eddie Reyes and Lin Zhong",

note = "Publisher Copyright: {\textcopyright} 2014 by the Association for Computing Machinery, Inc. (ACM).; 20th ACM Annual International Conference on Mobile Computing and Networking, MobiCom 2014 ; Conference date: 07-09-2014 Through 11-09-2014",

year = "2014",

month = sep,

day = "7",

doi = "10.1145/2639108.2642891",

language = "English (US)",

series = "Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM",

publisher = "Association for Computing Machinery",

pages = "387--389",

booktitle = "MobiCom 2014 - Proceedings of the 20th Annual",

}

TY - GEN

T1 - Poster

T2 - 20th ACM Annual International Conference on Mobile Computing and Networking, MobiCom 2014

AU - LiKamWa, Robert

AU - Reyes, Eddie

AU - Zhong, Lin

PY - 2014/9/7

Y1 - 2014/9/7

N2 - While computer vision algorithms and libraries have enabled and accelerated the adoption of vision processing into mobile and wearable applications, vision is a resource-hungry operation, and is thus not efficient enough to run on multiple applications simultaneously. However, we observe that many vision algorithms share identical sets of frames and features to perform their analyses, computed from the same library calls. Leveraging this observation, we design a split-process architecture to retrofit existing vision libraries to allow applications to transparently share the computational, memory, and energy overhead of vision processing.

AB - While computer vision algorithms and libraries have enabled and accelerated the adoption of vision processing into mobile and wearable applications, vision is a resource-hungry operation, and is thus not efficient enough to run on multiple applications simultaneously. However, we observe that many vision algorithms share identical sets of frames and features to perform their analyses, computed from the same library calls. Leveraging this observation, we design a split-process architecture to retrofit existing vision libraries to allow applications to transparently share the computational, memory, and energy overhead of vision processing.

KW - Remote execution

KW - Shared memory management

KW - Split-process

UR - http://www.scopus.com/inward/record.url?scp=84907822021&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84907822021&partnerID=8YFLogxK

U2 - 10.1145/2639108.2642891

DO - 10.1145/2639108.2642891

M3 - Conference contribution

AN - SCOPUS:84907822021

T3 - Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM

SP - 387

EP - 389

BT - MobiCom 2014 - Proceedings of the 20th Annual

PB - Association for Computing Machinery

Y2 - 7 September 2014 through 11 September 2014

ER -

Poster: Retrofitting computer vision libraries for concurrent support on mobile devices

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this