Fast Recursive Computation of Composite Correlation Filters
- 21 Downloads
Abstract—Algorithms for tracking multiple objects using correlation filters require calculation of a composite filter based on synthetic discriminant functions in order to increase the robustness to changes in posture, partial overlapping of objects by other objects, scaling, rotation, nonuniform illumination, and complex background. The calculation algorithm has large computational complexity. In this paper, we propose recursive calculation of a composite filter using algorithms for rapid inversion of matrices to accelerate synthesis of the filter. Computer simulation results on calculation of a composite correlation filter are presented and discussed in terms of the computation accuracy and the rate of calculation.
This work was supported by the Russian Science Foundation, grant no. 15-19-10010.
- 3.S. E. Ontiveros-Gallardo and V. Kober, “Objects tracking with adaptive correlation filters and kalman filtering,” Proc. SPIE’s 60 Ann. Meeting 9598, 95980X-8 (2015).Google Scholar
- 5.A. Ruchay and V. Kober, “A correlation-based algorithm for recognition and tracking of partially occluded objects,” Proc. SPIE 9971, 99712 (2016).Google Scholar
- 10.V. H. Díaz-Ramírez, K. Picos, and V. Kober, “Object tracking in nonuniform illumination using space-variant correlation filters,” in Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications (Proc. 18th Iberoamerican Congress, CIARP 2013, Havana, Cuba, Nov. 20–23, 2013), Ed. by J. Ruiz-Shulcloper and G. S. Di Baja, Part II, (Berlin, Springer-Verlag, 2013), pp. 455–462.Google Scholar
- 13.L. N. Gaxiola, V. H. Díaz-Ramírez, J. J. Tapia, et al., “Robust face tracking with locally-adaptive correlation filtering,” in Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications (Proc. 19th Iberoamerican Congress, CIARP 2014, Puerto Vallarta, Mexico, November 2–5, 2014, Ed. by E. Bayro-Corrochano and E. Hancock (Int. Springer, Cham, 2014), pp. 925–932.Google Scholar
- 17.R. Patnaik and D. Casasent, “Kernel synthetic discriminant function (sdf), filters for fast object recognition,” Proc. SPIE 7340, 7340–13 (2009).Google Scholar
- 19.E. M. Ramos-Michel and V. Kober, “Pattern recognition with an adaptive generalized sdf filter,” Proc. SPIE 6696, (2007).Google Scholar
- 20.R. A. Kerekes and B. V. K. V. Kumar, “Selecting a composite correlation filter design: a survey and comparative study,” Opt. Eng. 47, 47–18 (2008).Google Scholar
- 21.A. Ruchay and V. Kober, “Tracking of multiple objects with time-adjustable composite correlation filters,” Proc. SPIE 10396, 10396–6 (2017).Google Scholar