Complexity Reduction by Modified Scale-Space Construction in SIFT Generation Optimized for a Mobile GPU

Chulhee Lee; Chae Eun Rhee; Hyuk Jae Lee

doi:10.1109/TCSVT.2016.2580400

Complexity Reduction by Modified Scale-Space Construction in SIFT Generation Optimized for a Mobile GPU

Chulhee Lee, Chae Eun Rhee, Hyuk Jae Lee

Department of Electrical and Electronic Engineering

Research output: Contribution to journal › Article › peer-review

7 Citations (Scopus)

Abstract

Scale-invariant feature transform (SIFT) is one of the most widely used local features for computer vision in mobile devices. A mobile graphic processing unit (GPU) is often used to run computer-vision applications using SIFT features, but the performance in such a case is not powerful enough to generate SIFT features in real time. This paper proposes an efficient scheme to optimize the SIFT algorithm for a mobile GPU. It analyzes the conventional scale-space construction step in the SIFT generation, finding that reducing the size of the Gaussian filter and the scale-space image leads to a significant speedup with only a slight degradation of the quality of the features. Based on this observation, the SIFT algorithm is modified and implemented for real-time execution. Additional optimization techniques are employed for a further speedup by efficiently utilizing both the CPU and the GPU in a mobile processor. The proposed SIFT generation scheme achieves a processing speed of 28.30 frames/s for an image with a resolution of 1280 × 720 running on a Galaxy S5 LTE-A device, thereby gaining a speedup by the factors of 114.78 and 4.53 over CPU- and GPU-only implementations, respectively.

Original language	English
Article number	7490389
Pages (from-to)	2246-2259
Number of pages	14
Journal	IEEE Transactions on Circuits and Systems for Video Technology
Volume	27
Issue number	10
DOIs	https://doi.org/10.1109/TCSVT.2016.2580400
Publication status	Published - 2017 Oct

Bibliographical note

Funding Information:
Manuscript received August 17, 2015; revised December 5, 2015, February 11, 2016, and April 21, 2016; accepted June 10, 2016. Date of publication June 13, 2016; date of current version October 3, 2017. This work was supported in part by the Development of Fusion Processor based on Multi-Shader GPU within the Industrial Strategic Technology Development Program funded by the Ministry Of Trade, Industry & Energy under Grant 10041664; and in part by the Ministry of Science, ICT & Future Planning within the Information Technology Research Center supervised by the Institute for Information and Communications Technology Promotion under Grant IITP-2016-H8501-16-1005. This paper was recommended by Associate Editor L. Sousa.

Publisher Copyright:
© 2017 IEEE.

All Science Journal Classification (ASJC) codes

Media Technology
Electrical and Electronic Engineering

Access to Document

10.1109/TCSVT.2016.2580400

Cite this

@article{ce2de7e26bff4316be7bf6f694564a6e,

title = "Complexity Reduction by Modified Scale-Space Construction in SIFT Generation Optimized for a Mobile GPU",

abstract = "Scale-invariant feature transform (SIFT) is one of the most widely used local features for computer vision in mobile devices. A mobile graphic processing unit (GPU) is often used to run computer-vision applications using SIFT features, but the performance in such a case is not powerful enough to generate SIFT features in real time. This paper proposes an efficient scheme to optimize the SIFT algorithm for a mobile GPU. It analyzes the conventional scale-space construction step in the SIFT generation, finding that reducing the size of the Gaussian filter and the scale-space image leads to a significant speedup with only a slight degradation of the quality of the features. Based on this observation, the SIFT algorithm is modified and implemented for real-time execution. Additional optimization techniques are employed for a further speedup by efficiently utilizing both the CPU and the GPU in a mobile processor. The proposed SIFT generation scheme achieves a processing speed of 28.30 frames/s for an image with a resolution of 1280 × 720 running on a Galaxy S5 LTE-A device, thereby gaining a speedup by the factors of 114.78 and 4.53 over CPU- and GPU-only implementations, respectively.",

author = "Chulhee Lee and Rhee, {Chae Eun} and Lee, {Hyuk Jae}",

note = "Funding Information: Manuscript received August 17, 2015; revised December 5, 2015, February 11, 2016, and April 21, 2016; accepted June 10, 2016. Date of publication June 13, 2016; date of current version October 3, 2017. This work was supported in part by the Development of Fusion Processor based on Multi-Shader GPU within the Industrial Strategic Technology Development Program funded by the Ministry Of Trade, Industry & Energy under Grant 10041664; and in part by the Ministry of Science, ICT & Future Planning within the Information Technology Research Center supervised by the Institute for Information and Communications Technology Promotion under Grant IITP-2016-H8501-16-1005. This paper was recommended by Associate Editor L. Sousa. Publisher Copyright: {\textcopyright} 2017 IEEE.",

year = "2017",

month = oct,

doi = "10.1109/TCSVT.2016.2580400",

language = "English",

volume = "27",

pages = "2246--2259",

journal = "IEEE Transactions on Circuits and Systems for Video Technology",

issn = "1051-8215",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "10",

}

TY - JOUR

T1 - Complexity Reduction by Modified Scale-Space Construction in SIFT Generation Optimized for a Mobile GPU

AU - Lee, Chulhee

AU - Rhee, Chae Eun

AU - Lee, Hyuk Jae

N1 - Funding Information: Manuscript received August 17, 2015; revised December 5, 2015, February 11, 2016, and April 21, 2016; accepted June 10, 2016. Date of publication June 13, 2016; date of current version October 3, 2017. This work was supported in part by the Development of Fusion Processor based on Multi-Shader GPU within the Industrial Strategic Technology Development Program funded by the Ministry Of Trade, Industry & Energy under Grant 10041664; and in part by the Ministry of Science, ICT & Future Planning within the Information Technology Research Center supervised by the Institute for Information and Communications Technology Promotion under Grant IITP-2016-H8501-16-1005. This paper was recommended by Associate Editor L. Sousa. Publisher Copyright: © 2017 IEEE.

PY - 2017/10

Y1 - 2017/10

N2 - Scale-invariant feature transform (SIFT) is one of the most widely used local features for computer vision in mobile devices. A mobile graphic processing unit (GPU) is often used to run computer-vision applications using SIFT features, but the performance in such a case is not powerful enough to generate SIFT features in real time. This paper proposes an efficient scheme to optimize the SIFT algorithm for a mobile GPU. It analyzes the conventional scale-space construction step in the SIFT generation, finding that reducing the size of the Gaussian filter and the scale-space image leads to a significant speedup with only a slight degradation of the quality of the features. Based on this observation, the SIFT algorithm is modified and implemented for real-time execution. Additional optimization techniques are employed for a further speedup by efficiently utilizing both the CPU and the GPU in a mobile processor. The proposed SIFT generation scheme achieves a processing speed of 28.30 frames/s for an image with a resolution of 1280 × 720 running on a Galaxy S5 LTE-A device, thereby gaining a speedup by the factors of 114.78 and 4.53 over CPU- and GPU-only implementations, respectively.

AB - Scale-invariant feature transform (SIFT) is one of the most widely used local features for computer vision in mobile devices. A mobile graphic processing unit (GPU) is often used to run computer-vision applications using SIFT features, but the performance in such a case is not powerful enough to generate SIFT features in real time. This paper proposes an efficient scheme to optimize the SIFT algorithm for a mobile GPU. It analyzes the conventional scale-space construction step in the SIFT generation, finding that reducing the size of the Gaussian filter and the scale-space image leads to a significant speedup with only a slight degradation of the quality of the features. Based on this observation, the SIFT algorithm is modified and implemented for real-time execution. Additional optimization techniques are employed for a further speedup by efficiently utilizing both the CPU and the GPU in a mobile processor. The proposed SIFT generation scheme achieves a processing speed of 28.30 frames/s for an image with a resolution of 1280 × 720 running on a Galaxy S5 LTE-A device, thereby gaining a speedup by the factors of 114.78 and 4.53 over CPU- and GPU-only implementations, respectively.

UR - http://www.scopus.com/inward/record.url?scp=85032296492&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85032296492&partnerID=8YFLogxK

U2 - 10.1109/TCSVT.2016.2580400

DO - 10.1109/TCSVT.2016.2580400

M3 - Article

AN - SCOPUS:85032296492

SN - 1051-8215

VL - 27

SP - 2246

EP - 2259

JO - IEEE Transactions on Circuits and Systems for Video Technology

JF - IEEE Transactions on Circuits and Systems for Video Technology

IS - 10

M1 - 7490389

ER -

Complexity Reduction by Modified Scale-Space Construction in SIFT Generation Optimized for a Mobile GPU

Abstract

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this