NDLI: Performance optimization of vision apps on mobile application processor

Content Provider	IEEE Xplore Digital Library
Author	Kwang-Ting Cheng Xin Yang Yi-Chu Wang
Copyright Year	2013
Description	Author affiliation: Dept. of Electr. & Comput. Eng., Univ. of California, Santa Barbara, Santa Barbara, CA, USA (Kwang-Ting Cheng; Xin Yang; Yi-Chu Wang)
Abstract	Optimizing performance of compute-intensive vision apps running on mobile application processor (AP) is critical to satisfactory experience for smartphone and tablet users. Most existing vision algorithms were primarily designed and implemented for desktop and server platforms. Porting them to a mobile platform without adapting the algorithms to account for the platform's limitations would cause serious algorithmhardware mismatches, yielding unnecessary runtime degradation. Modern mobile AP, which integrates multicore CPUs, GPUs and other special-purpose accelerators, offers multiple options of porting vision apps to various computing cores. To develop an optimized implementation for a vision app, it is necessary to understand the potential mismatches for better algorithm adaptation and optimized mapping of the algorithm to a handheld platform. In this paper, we identify mismatches and propose adaptation guidelines for three different porting strategies: porting an algorithm to 1) a mobile CPU, 2) a mobile GPU, and 3) mobile heterogeneous multicores (mobile CPUs+GPUs). For each strategy, we illustrate the adaptation/porting guidelines using an exemplar vision task. Experimental results demonstrate that with proper adaptation following the proposed guidelines, we could achieve a significant speedup with little accuracy drop.
Starting Page	187
Ending Page	191
File Size	728599
Page Count	5
File Format	PDF
ISBN	9781479909414
ISSN	21578672
e-ISBN	9781479909445
DOI	10.1109/IWSSIP.2013.6623485
Language	English
Publisher	Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher Date	2013-07-07
Publisher Place	Romania
Access Restriction	Subscribed
Rights Holder	Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subject Keyword	algorithm adaption Runtime Face recognition Graphics processing units Mobile communication Feature extraction heterogeneous parallel computing mobile application processor Acceleration Face GPU acceleration
Content Type	Text
Resource Type	Article

Sl.	Authority	Responsibilities	Communication Details
1	Ministry of Education (GoI), Department of Higher Education	Sanctioning Authority	https://www.education.gov.in/ict-initiatives
2	Indian Institute of Technology Kharagpur	Host Institute of the Project: The host institute of the project is responsible for providing infrastructure support and hosting the project	https://www.iitkgp.ac.in
3	National Digital Library of India Office, Indian Institute of Technology Kharagpur	The administrative and infrastructural headquarters of the project	Dr. B. Sutradhar bsutra@ndl.gov.in
4	Project PI / Joint PI	Principal Investigator and Joint Principal Investigators of the project	Dr. B. Sutradhar bsutra@ndl.gov.in Prof. Saswat Chakrabarti will be added soon
5	Website/Portal (Helpdesk)	Queries regarding NDLI and its services	support@ndl.gov.in
6	Contents and Copyright Issues	Queries related to content curation and copyright issues	content@ndl.gov.in
7	National Digital Library of India Club (NDLI Club)	Queries related to NDLI Club formation, support, user awareness program, seminar/symposium, collaboration, social media, promotion, and outreach	clubsupport@ndl.gov.in
8	Digital Preservation Centre (DPC)	Assistance with digitizing and archiving copyright-free printed books	dpc@ndl.gov.in
9	IDR Setup or Support	Queries related to establishment and support of Institutional Digital Repository (IDR) and IDR workshops	idr@ndl.gov.in

A fast and efficient sift detector using the mobile GPU

Runtime Adaptation for Autonomic Heterogeneous Computing

Energy and Performance Characterization of Mobile Heterogeneous Computing

Parallel computing of WaveCluster algorithm for face recognition application

A low power and high performance face detection on mobile GPU

Accelerating local feature extraction using OpenCL on heterogeneous platforms

Accelerating computer vision algorithms using OpenCL framework on the mobile GPU - A case study

Using mobile GPU for general-purpose computing – a case study of face recognition on smartphones

GPU-ASIFT: A fast fully affine-invariant feature extraction algorithm

Performance optimization of vision apps on mobile application processor

Similar Documents

A fast and efficient sift detector using the mobile GPU

Runtime Adaptation for Autonomic Heterogeneous Computing

Energy and Performance Characterization of Mobile Heterogeneous Computing

Parallel computing of WaveCluster algorithm for face recognition application

A low power and high performance face detection on mobile GPU

Accelerating local feature extraction using OpenCL on heterogeneous platforms

Accelerating computer vision algorithms using OpenCL framework on the mobile GPU - A case study

Using mobile GPU for general-purpose computing – a case study of face recognition on smartphones

GPU-ASIFT: A fast fully affine-invariant feature extraction algorithm

Performance optimization of vision apps on mobile application processor