A Fast and Efficient Sift Detector Using The Mobile GPU

Simple item page
dc.citation.conferenceDate2013en_US
dc.citation.conferenceNameIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)en_US
dc.citation.firstpage2674
dc.citation.lastpage2678
dc.citation.locationVancouver, Canadaen_US
dc.contributor.authorRister, Blaine
dc.contributor.authorWang, Guohui
dc.contributor.authorWu, Michael
dc.contributor.authorCavallaro, Joseph R.
dc.date.accessioned2013-10-25T21:19:47Z
dc.date.available2013-10-25T21:19:47Z
dc.date.issued2013-06
dc.description.abstractEmerging mobile applications, such as augmented reality, demand robust feature detection at high frame rates. We present an implementation of the popular Scale-Invariant Feature Transform (SIFT) feature detection algorithm that incorporates the powerful graphics processing unit (GPU) in mobile devices. Where the usual GPU methods are inefficient on mobile hardware, we propose a heterogeneous dataflow scheme. By methodically partitioning the computation, compressing the data for memory transfers, and taking into account the unique challenges that arise out of the mobile GPU, we are able to achieve a speedup of 4-7x over an optimized CPU version, and a 6.4x speedup over a published GPU implementation. Additionally, we reduce energy consumption by 87 percent per image. We achieve near-realtime detection without compromising the original algorithm.en_US
dc.description.sponsorshipSamsungen_US
dc.description.sponsorshipUS National Science Foundation grant EECS- 1232274en_US
dc.description.sponsorshipUS National Science Foundation grant EECS-0925942en_US
dc.description.sponsorshipUS National Science Foundation grant CNS-0923479en_US
dc.identifier.citationB. Rister, G. Wang, M. Wu and J. R. Cavallaro, "A Fast and Efficient Sift Detector Using The Mobile GPU," 2013.*
dc.identifier.doihttp://dx.doi.org/10.1109/ICASSP.2013.6638141en_US
dc.identifier.urihttps://hdl.handle.net/1911/75008
dc.language.isoengen
dc.publisherIEEEen_US
dc.subjectcomputer visionen_US
dc.subjectmobile computingen_US
dc.subjectfeature detectionen_US
dc.subjectgraphics processing unit (GPU)en_US
dc.subjectOpenGL for Embedded Systems (OpenGL ES)en_US
dc.titleA Fast and Efficient Sift Detector Using The Mobile GPUen_US
dc.typeConference paperen_US
dc.type.dcmiTexten
dc.type.dcmiTexten_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
2013_ICASSP_Rister_0002674.pdf
Size:
320.8 KB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.61 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
ECE Publications