Parallelization and Optimization of SIFT Feature Extraction on Cluster System

Authors: Mingling Zheng, Zhenlong Song, Ke Xu, Hengzhu Liu

Abstract:

Scale Invariant Feature Transform (SIFT) has been widely applied, but extracting SIFT feature is complicated and time-consuming. In this paper, to meet the demand of the real-time applications, SIFT is parallelized and optimized on cluster system, which is named pSIFT. Redundancy storage and communication are used for boundary data to improve the performance, and before representation of feature descriptor, data reallocation is adopted to keep load balance in pSIFT. Experimental results show that pSIFT achieves good speedup and scalability.

Keywords: cluster, image matching, parallelization and optimization, SIFT.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1084135

References:

[1] D. G. Lowe, "Distinctive image features from scale-invariant keypoints," International Journal of Computer Vision, vol. 60, pp. 91-110, 2004.
[2] A.Y.Ke and R.Sukthankar, "PCA-SIFT: A more distinctive representation for local image descriptors," In Proc. 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR-04), pp.506-513.
[3] Mikolajczyk, K., Schmid, C., "A performance evaluation of local descriptors," IEEE Trans. Pattern Analysis and Machine Intelligence. Vol.27, pp.1615-1630, Augst 2005.
[4] Alaa E. Abdel-Hakim and Aly A. Farag, "CSIFT: A SIFT Descriptor with Color Invariant Characteristics," in proc. 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR-06).
[5] Geoffrey Treen, and Anthony Whitehead, "Efficient SIFT Matching from Keypoint Descriptor Properties," 2009 Workshop on Applications of Computer Vision(WACV), pp1-7.
[6] Vanderlei Bonato, Eduardo Marques, and George A. Constantinides, "A Parallel Hardware Architecture for Scale and Rotation Invariant Feature Detection," IEEE Trans. Circuits and Systems for Video Technology, VOL.18, pp1703-1712, 2008.
[7] Seth Warn, Wesley Emeneker , Jackson Cothren,Amy Apon, "Accelerating SIFT on Parallel Architectures," In Proc. 2009 IEEE Int. Conf. Cluster Computing and Workshops(CLUSTER-09), pp.1-4.
[8] Marc Lalonde, David Byrns, Langis Gagnon, Normand Teasdale, Denis Laurendeau, "Real-time eye blink detection with GPU-based SIFT tracking," In Proc. 4th Canadian Conference on Computer and Robot Vision(CRV'07), pp.481-487,2007.
[9] Sirmacek, B., Unsalan, C., "Urban-Area and Building Detection Using SIFT Keypoints and Graph Theory," IEEE Trans. Geoscience and Remote Sensing, Vol.47, pp.1156-1167, 2009.
[10] Gangqiang Zhao, Ling Chen, Jie Song, Gencai Chen, "Large head movement tracking using SIFT-based registration," In Proc. 15th international conference on Multimedia, PP: 807-810, 2007.
[11] Jiang, R.M., Crookes, D., Luo, N., Davidson, M.W., "Live-Cell Tracking Using SIFT Features in DIC Microscopic Videos," IEEE Trans. Biomedical Engineering, Vol.57, pp: 2219-2228, 2010.
[12] Goncalves, H., Corte-Real, L., Goncalves, J.A., "Automatic Image Registration through Image Segmentation and SIFT," IEEE Trans. Geoscience and Remote Sensing, Vol.49 pp.2589-2600, 2011.
[13] Yi, Z., Zhiguo, C., Yang, X., "Multi-spectral remote image registration based on SIFT," IEEE Electronics Letters, Vol.44 pp. 107-108,2008.
[14] Sudipta N. Sinha, Jan-Michael Frahm, Marc Pollefeys, and Yakup Genc, "Feature Tracking and Matching in Video Using Programmable Graphics Hardware," Machine Vision and Applications, Vol.22, pp.207-217, March 2007.
[15] S. Heymann, K. Muller, A. Smolic, B. Froehlich, and T. Wiegand, "SIFT implementation and optimization for general-purpose GPU," In Proc. WSCG-07, 2007.
[16] Q. Zhang, Y. Chen, Y. Zhang, and Y. Xu, "Sift implementation and optimization for multi-core systems," IEEE International Symposium on Parallel and Distributed Processing (IPDPS 2008), pp. 1-8, 2008.
[17] H. Feng, E. Li, Y. Chen, and Y. Zhang, "Parallelization and characterization of sift on multi-core systems," IEEE International Symposium on Workload Characterization (IISWC-08), pp. 14-23, 2008.
[18] Guiyuan Jiang, Guiling Zhang and Dakun Zhang, "A Distributed Dynamic Parallel Algorithm for SIFT Feature Extraction," 3rd International Symposium on Parallel Architectures, Algorithms and Programming (PAAP), pp.381-385, 2010.
[19] Lowe, D.G., "Object recognition from local scale-invariant features," In Proc. IEEE Int Conf. Computer Vision, pp. 1150-1157, 1999.
[20] Andrea Vedaldi, SIFT source code, download from http://www.vlfeat.org/~vedaldi/assets/ siftpp/versions/.