Abstract
We present a faster than real-time parallel implementation of standard sum of squared differences (SSD) stereo vision algorithm, on an SIMD architecture, the CSX700. To our knowledge, this is the first highly parallel implementation of this algorithm using 192 processing elements. For disparity range of 16 pixels, we have achieved the rate of 160 and 59 stereo pairs per second on 640x480 and 1280x720 images, respectively. Since this implementation is much faster than real time, it leaves enough time for performing other machine vision applications in real time. Our results demonstrate that CSX architecture is a powerful processor for (low level) computer vision applications. Due to the low-power consumption of CSX architecture, it can be a good candidate for mobile computer vision applications.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Scharstein, D., Szeliski, R.: A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. International Journal of Computer Vision 47, 7–42 (2001)
Hirschmüller, H., Innocent, P.R., Garibaldi, J.: Real-time correlation-based stereo vision with reduced border errors. Int. J. Comput. Vision 47, 229–246 (2002)
Sunyoto, H., van der Mark, W., Gavrila, D.M.: A comparative study of fast dense stereo vision algorithms. In: 2004 IEEE Intelligent Vehicles Symposium, pp. 319–324 (2004)
Yang, R., Pollefeys, M.: A versatile stereo implementation on commodity graphics hardware. Real-Time Imaging 11, 7–18 (2005)
Wang, L., Liao, M., Gong, M., Yang, R., Nister, D.: High-quality real-time stereo using adaptive cost aggregation and dynamic programming. In: 3DPVT 2006: Proceedings of the Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT 2006), Washington, DC, USA, pp. 798–805. IEEE Computer Society, Los Alamitos (2006)
Woodfill, J., Herzen, B.V.: Real-time stereo vision on the parts reconfigurable computer. In: IEEE Symposium on FPGAs for Custom Computing Machines, pp. 242–250. IEEE Computer Society Press, Los Alamitos (1997)
Murphy, C., Lindquist, D., Rynning, A.M., Cecil, T., Leavitt, S., Chang, M.L.: Low-cost stereo vision on an fpga. In: FCCM 2007: Proceedings of the 15th Annual IEEE Symposium on Field-Programmable Custom Computing Machines, Washington, DC, USA, pp. 333–334. IEEE Computer Society, Los Alamitos (2007)
ClearSpeed (2009), http://www.clearspeed.com
ClearSpeed Technology: ClearSpeed Whitepaper: CSX Processor Architecture (2007), http://www.clearspeed.com
Baldonado, M., Chang, C.C., Gravano, L., Paepcke, A.: The stanford digital library metadata architecture. Int. J. Digit. Libr. 1, 108–121 (1997)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Hosseini, F., Fijany, A., Safari, S., Chellali, R., Fontaine, JG. (2009). Real-Time Parallel Implementation of SSD Stereo Vision Algorithm on CSX SIMD Architecture. In: Bebis, G., et al. Advances in Visual Computing. ISVC 2009. Lecture Notes in Computer Science, vol 5875. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-10331-5_75
Download citation
DOI: https://doi.org/10.1007/978-3-642-10331-5_75
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-10330-8
Online ISBN: 978-3-642-10331-5
eBook Packages: Computer ScienceComputer Science (R0)