Abstract
The depth map extraction from stereo video is a key technology of stereoscopic 3D video as well as view synthesis and 2D-3D video conversions. Sum of Absolute Differences (SAD) is a representative method to reconstruct disparity map. However, dense disparity map implementation requires heavy computations and extensive memory accesses. In this situation, the rapid advance of computer hardware and popularity of multimedia applications enable multi-core processors to become a dominant market trend in desk-top PCs as well as high end mobile devices; this movement allows many parallel programming technologies to be realized in users computing devices. Therefore, this paper proposes a parallel algorithm for SAD operation using a shared memory programming, OpenMP, which can provide the advantage to simplify managing and synchronization of program threads. The parallel implementation results show the 2.5 times of performance improvements on the processing speed compared with the serial implementation.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Kayi A, Yao Y, El-Ghazawi T, Newby G (2007) Experimental evaluation of emerging multi-core architectures. In: Proceeding of IPDPS 2007, pp 1–6
Kim CG, Kim JG, Lee DH Optimizing image processing on multi-core CPUs with Intel parallel programming technologies. Multimedia tools applications, Springer. doi:10.1007/s11042-011-0906-y
OpenMP official web site: www.openmp.org
Blume H, von Livonius J, Rotenberg L, Noll TG, Bothe H, Brakensiek J (2008) OpenMP-based parallelization on an MPCore multiprocessor platform—a performance and power analysis. J Syst Architect 54:1019–1029
Scharstein D, Szeliski R (2002) A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. Int J Comput Vis 47:7–42
Kamencay P, Breznan M, Jarina R, Lukac P, Zachariasova M (2012) Improved depth map estimation from stereo images based on hybrid method. Radioengineering 21(1):70–78
Kuhl A (2005) Comparison of stereo matching algorithms for mobile robots. Centre for Intelligent Information Processing System, University of Western Australia, pp 4–24
Song Y, Ho YS (2011) Fast disparity map estimation using multi-thread parallel processing. In: International Conference on Embedded Systems and Intelligent Technology (ICESIT), pp 1–4
Acknowledgments
This research is supported by Ministry of Culture, Sports and Tourism (MCST) and Korea Creative Content Agency (KOCCA) in the Culture Technology (CT) Research and Development Program 2013.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer Science+Business Media Dordrecht
About this paper
Cite this paper
Kim, C.G. (2013). Parallel SAD for Fast Dense Disparity Map Using a Shared Memory Programming. In: Park, J., Barolli, L., Xhafa, F., Jeong, HY. (eds) Information Technology Convergence. Lecture Notes in Electrical Engineering, vol 253. Springer, Dordrecht. https://doi.org/10.1007/978-94-007-6996-0_112
Download citation
DOI: https://doi.org/10.1007/978-94-007-6996-0_112
Published:
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-007-6995-3
Online ISBN: 978-94-007-6996-0
eBook Packages: EngineeringEngineering (R0)