Advanced

Fast GPU Accelerated Stereo Correspondence for Embedded Surveillance Camera Systems

Andersson Sturén, Vilhelm LU and Hindefelt, Sebastian (2015) In LU-CS-EX 2015-33 EDA920 20151
Department of Computer Science
Abstract (Swedish)
Many surveillance applications could benefit from the use of stereo cam-
eras for depth perception. While state-of-the-art methods provide high quality scene depth information, many of the methods are very time consuming and not suitable for real-time usage in limited embedded systems. This study was conducted to examine stereo correlation methods to find a suitable algorithm for real-time or near real-time depth perception through disparity maps in a stereo video surveillance camera with an embedded GPU. Moreover, novel refinements and alternations was investigated to further improve performance and quality. Quality tests were conducted in Octave while GPU suitability and performance tests were done in C++ with the OpenGL ES 2.0 library.... (More)
Many surveillance applications could benefit from the use of stereo cam-
eras for depth perception. While state-of-the-art methods provide high quality scene depth information, many of the methods are very time consuming and not suitable for real-time usage in limited embedded systems. This study was conducted to examine stereo correlation methods to find a suitable algorithm for real-time or near real-time depth perception through disparity maps in a stereo video surveillance camera with an embedded GPU. Moreover, novel refinements and alternations was investigated to further improve performance and quality. Quality tests were conducted in Octave while GPU suitability and performance tests were done in C++ with the OpenGL ES 2.0 library. The result is a local stereo correlation method using Normalized Cross Correlation together with sparse support windows and a suggested improvement for pixel-wise matching confidence. Applying sparse support windows increased frame rate by 35% with minimal quality penalty as compared to using full support windows. (Less)
Please use this url to cite or link to this publication:
author
Andersson Sturén, Vilhelm LU and Hindefelt, Sebastian
supervisor
organization
course
EDA920 20151
year
type
H3 - Professional qualifications (4 Years - )
subject
keywords
disparity map, 3D vision, camera surveillance, images and graphics, surveillance, Stereo correspondence, GPU accelerated, fast
publication/series
LU-CS-EX 2015-33
report number
LU-CS-EX 2015-33
ISSN
1650-2884
language
English
id
7761572
date added to LUP
2015-08-14 10:29:18
date last changed
2015-08-14 10:29:18
@misc{7761572,
  abstract     = {Many surveillance applications could benefit from the use of stereo cam-
eras for depth perception. While state-of-the-art methods provide high quality scene depth information, many of the methods are very time consuming and not suitable for real-time usage in limited embedded systems. This study was conducted to examine stereo correlation methods to find a suitable algorithm for real-time or near real-time depth perception through disparity maps in a stereo video surveillance camera with an embedded GPU. Moreover, novel refinements and alternations was investigated to further improve performance and quality. Quality tests were conducted in Octave while GPU suitability and performance tests were done in C++ with the OpenGL ES 2.0 library. The result is a local stereo correlation method using Normalized Cross Correlation together with sparse support windows and a suggested improvement for pixel-wise matching confidence. Applying sparse support windows increased frame rate by 35% with minimal quality penalty as compared to using full support windows.},
  author       = {Andersson Sturén, Vilhelm and Hindefelt, Sebastian},
  issn         = {1650-2884},
  keyword      = {disparity map,3D vision,camera surveillance,images and graphics,surveillance,Stereo correspondence,GPU accelerated,fast},
  language     = {eng},
  note         = {Student Paper},
  series       = {LU-CS-EX 2015-33},
  title        = {Fast GPU Accelerated Stereo Correspondence for Embedded Surveillance Camera Systems},
  year         = {2015},
}