Abstract
Motion object tracking is one of the most important research directions in computer vision. Challenges in designing a tracking method are usually caused by occlusions, noise, or illumination changes. In this paper, a robust visual tracking algorithm is proposed in order to cope with the occlusion by introducing the motion object tracking issue as a low-rank matrix representation problem. First, being the main contribution of this paper, the observation matrix composed by image sequences is decomposed into a low-rank matrix and a sparse matrix. The motion object in the image sequence forms the low-rank matrix and the occlusion on the motion object forms the sparse matrix. Then the motion object tracking is carried out using a Bayesian state under the particle filter framework. Finally, an effective alternating algorithm is utilized to solve the proposed optimization formulation. The proposed algorithm has been examined throughout several challenging image sequences, and experiment results show that it works effectively and efficiently in different situations.
Similar content being viewed by others
References
Wax, N.: Signal to noise improvement and the statistics of tracking populations. J. Appl. Phys. 26(5), 586–595 (1995)
Li, X., Hu, W., Shen, C., et al.: A survey of appearance models in visual object tracking. ACM Trans. Intell. Syst. Technol. 4(4), 478–488 (2013)
Davis, L.S., Harwood, D., Haritaoglu, I.: W4: real-time surveillance of people and their activities. IEEE Trans. Pattern Anal. Mach. Intell. 22(8), 809–830 (2000)
Piscataway, N.: Multi-camera multi-person tracking for easy living. In: IEEE International Workshop on Visual Surveillance, pp. 3–10 (2000)
Coifman, B., Beymer, D., Mclauchlan, P., et al.: A real-time computer vision system for vehicle tracking and traffic surveillance. Transport. Res. Part C Emerg. Technol. 6(4), 271–288 (1998)
Tai, J., Tsang, S., Lin, C., Song, K.: Real-time image tracking for automatic traffic monitoring and enforcement application. Image Vis. Comput. 22(6), 485–501 (2004)
Masoud, O., Papani, N.: A novel method for tracking and counting pedestrians in real time using a single camera. IEEE Trans. Veh. Technol. 50(5), 1267–1278 (2001)
Pavlovie, V.I., Sharma, R., Huang, T.: Visual interpretation of hand gestures for human-computer interaction: a review. IEEE Trans. Pattern Anal. Mach. Intell. 19(7), 677–695 (1997)
Paschalakis, S., Bober, M.: Real-time face detection and tracking for mobile videoconferencing. Real Time Imaging 10(2), 81–94 (2004)
Yilmaz, A., Javed, O., Shah, M.: Object tracking: a survey. ACM Comput. Surv. 38(4), 1 (2006)
Wright, J., Yang, A.Y., Ganesh, A., Sastry, S.S., Ma, Y.: Robust face recognition via sparse representation. PAMI 31(1), 210–227 (2009)
Zhou, X., Yang, C., Yu, W.: Moving object detection by detecting contiguous outliers in the low-rank representation. Pattern Anal. Mach. Intell. IEEE Trans. 35(3), 597–610 (2013)
Mazumder, R., Hastie, T., Tibshirani, R.: Spectral regularization algorithms for learning large incomplete matrices. J. Mach. Learn. Res. 11, 2287–2322 (2010)
Stan Z.L.: Markov random field modeling in image analysis. In: Markov Random Field Modeling in Image Analysis, p. xxiv + 357 (1995)
Salti, S., Cavallaro, A., Stefano, L.D.: Adaptive appearance modeling for video tracking: survey and evaluation. IEEE Trans. Image Process. 21(10), 4334–4348 (2012)
Mei, X., Ling, H.: Robust visual tracking using L1 minimization. In: IEEE 12th International Conference on Computer Vision, 2009, pp. 1436–1443. IEEE (2009)
Comaniciu, D., Ramesh, V., Meer, P.: Kernel-based object tracking. PAMI 25, 564–577 (2003)
Doucet, A., Freitas, J.F.G.D., Gordon, N.J., et al. (eds.): Sequential Monte Carlo Methods in Practice. Springer, New York (2000)
Isard, M., Blake, A.: Condensation-conditional density propagation for visual tracking. IJCV 29(2), 5–28 (1998)
Yang, C., Duraiswami, R., Davis, L.: Fast multiple object tracking via a hierarchical particle filter. In: Computer Vision, 2005. ICCV 2005. Tenth IEEE International Conference, vol. 1, pp. 212–219. IEEE (2005)
Zhou, S.K., Chellappa, R., Moghaddam, B.: Visual tracking and recognition using appearance-adaptive models in particle filters. IEEE Trans. Image Process. 11(2), 1491–1506 (2004)
Wright, J., Ma, Y., Mairal, J., Sapiro, G., Huang, T., Yan, S.: Sparse representation for computer vision and pattern recognition. Proc. IEEE 98(6), 1031–1044 (2010)
Peng, Y., Ganesh, A., Wright, J., Xu, W., Ma, Y.: RASL: robust alignment by sparse and low-rank decomposition for linearly correlated images. IEEE Trans. Pattern Anal. Mach. Intell. 34(11), 2233–2246 (2011)
Mei, X., Ling, H.: Robust visual tracking and vehicle classification via sparse representation. IEEE Trans. Pattern Anal. Mach. Intell. 33(11), 2259–2272 (2011)
Liu, B., Yang, L., Huang, J., et al. Robust and fast collaborative tracking with two stage sparse optimization. In: Lecture Notes in Computer Science, pp. 624–637 (2010)
Jia, X., Lu, H., Yang, M.H.: Visual tracking via adaptive structural local sparse appearance model. In: Computer vision and pattern recognition (CVPR), 2012 IEEE Conference, pp. 1822–1829. IEEE (2012)
Szeliski, R.: Computer vision: Algorithms and Applications. Springer Science & Business Media, New York (2010)
Oliver, N., Rosario, B., Pentland, A.: A Bayesian computer vision system for modeling human interactions. IEEE Trans. Pattern Anal. Mach. Intell. 22(8), 831–843 (2000)
Geman, S., Geman, D.: Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images. IEEE Trans. Pattern Anal. Mach. Intell. 6(6), 721–741 (1984)
Recht, B., Fazel, M., Parrilo, P.: Guaranteed minimum-rank solutions of linear matrix equations via nuclear norm minimization. SIAM Rev. 52(3), 471–501 (2010)
Matthews, I., Ishikawa, T., Baker, S.: The template update problem. PAMI 4, 810–815 (2004)
Kaneko, T., Hori, O.: Feature selection for reliable tracking using template matching. In: 2013 IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, p. 796 (2003)
Peng, C., Chen, Q., Qian, W.X.: Eigen space-based tracking for feature points. Opt. Rev. 21(3), 304–312 (2014)
Ross, D.A., Lim, J., Lin, R.S., et al.: Incremental learning for robust visual tracking. Int. J. Comput. Vis. 77(1–3), 125–141 (2008)
Wu, Y., Lim, J., Yang, M.H.: Online object tracking: a benchmark. In: Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference, pp. 2411–2418. IEEE (2013)
Babenko, B., Yang, M.H., Belongie, S.: Robust object tracking with online multiple instance learning. Pattern Anal. Mach. Intell. IEEE Trans. 33(8), 1619–1632 (2011)
Everingham, M., Van Gool, L., Williams, C.K.I., et al.: The pascal visual object classes (voc) challenge. Int. J. Comput. Vis. 88(2), 303–338 (2010)
Mutelo, R.M., Woo, W.L., Dlay, S.S.: Two dimensional principle component analysis of Gabor features for face representation and recognition. In: Communication Systems, Networks and Digital Signal Processing, 2008. CNSDSP 2008. 6th International Symposium. IEEE, pp. 457–461 (2008)
Cai, J.F., Candes, E.J., Shen, Z.: A singular value thresholding algorithm for matrix completion. SIAM J. Optim. 20(4), 1956–1982 (2010)
Boykov, Y., Veksler, O., Zabih, R.: Fast approximate energy minimization via graph cuts. IEEE Trans. Pattern Anal. Mach. Intell. 23(11), 1222–1239 (2001)
Kolmogorov, V., Zabih, R.: What energy functions can be minimized via graph cuts? IEEE Trans. Pattern Anal. Mach. Intell. 26(2), 147–159 (2004)
Acknowledgments
This project is supported by the Natural Science Foundation of Jiangsu Province of China, under Grant No. BK20130769; Jiangsu Province High-level Talents in Six Industries, No. 2012-DZXX-037; and Program for New Century Excellent Talents in University, No. NCET-12-0630.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
Formula (13) is nonconvex and it includes both discrete and continuous variable. It is extremely difficult to solve for the solutions of \(T\) and \(S\). Hence, we can adopt an alternating algorithm that separates formula (13) over \(T\) and \(S\) into two steps to solve this problem.
1.1 Estimation of the low-rank matrix \(T\)
Given an estimate of the occlusion support matrix \(\hat{S}\), the minimization in formula (13) over \(T\) turns out to be the matrix completion problem [13]:
The optimal \(T\) in formula (20) can be computed efficiently through the algorithm of SOFT-IMPUTE [30], which makes use of the lemma proposed in Ref. [39]. The solution is given as:
where \(\varTheta_{\alpha }\) means the singular value thresholding
where \(\varSigma_{\alpha } = {\text{diag}}[\left( {d_{1} - \alpha } \right),\left( {d_{2} - \alpha } \right), \ldots ,\left( {d_{r} - \alpha } \right)]\), \(U\varSigma V^{T}\) is the SVD of \({\mathbf{I}}_{1:n}\), and \(\varSigma = {\text{diag}}[d_{1} ,d_{2} , \ldots ,d_{r} ]\).
1.2 Estimation of the occlusion support matrix \(S\)
The energy in formula (12) can be rewritten as follows:
where \(C = \frac{1}{2}\sum\nolimits_{it} {\left( {{\mathbf{I}}_{1:n} - \hat{T}} \right)^{2} }\) is a constant when \(\hat{T}\) is fixed. The above energy is in the standard form of the first-order MRFs with binary labels [40], which can be solved exactly using graph cuts [41].
Rights and permissions
About this article
Cite this article
Kong, X., Chen, Q., Xu, F. et al. Motion object tracking based on the low-rank matrix representation. Opt Rev 22, 786–801 (2015). https://doi.org/10.1007/s10043-015-0131-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10043-015-0131-4