Abstract
The covariance descriptor has received an increasing amount of interest in visual tracking. However, the conventional covariance tracking algorithms fail to estimate both the scale and orientation of an object. In this paper, we present a kernel-based region covariance descriptor to address this issue. An affine kernel function is incorporated to the covariance matrix to effectively control the correlations among extracted features inside the object region. Under the Log-Euclidean Riemannian metric, we construct a region similarity measure function that describes the relationship between the candidate and a given appearance template. The tracking task is then implemented by minimizing the similarity measure, in which the gradient descent method is utilized to iteratively optimize affine transformation parameters. In addition, the template is dynamically updated by computing the geometric mean of covariance matrices in Riemannian manifold for adapting to the appearance changes of the object over time. Experimental results compared with several relevant tracking methods demonstrate the good performance of the proposed algorithm under challenging conditions.
Similar content being viewed by others
Notes
We manually selected target location in the first frame. Then, the covariance feature \(\varvec{C}_{T}\) extracted from the selected image is considered as the template.
The test sequences are obtained from http://visual-tracking.net/#.
References
Li X, Hu W, Shen C, Zhang Z, Dick A, Hengel AVD (2013) A survey of appearance models in visual object tracking. ACM Trans Intell Syst Technol (TIST) 4(4):58
Adam A, Rivlin E, Shimshoni I (2006) Robust fragments-based tracking using the integral histogram. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR’06), vol 1, pp 798–805
Wu Y, Pei M, Yang M, He Y, Jia Y (2014) Landmark-based inductive model for robust discriminative tracking. In: Proceedings of the 12th Asian conference on computer vision (ACCV2014), in press
Ross D, Lim J, Lin R, Yang M (2008) Incremental learning for robust visual tracking. Int J Comput Vis 77(1):125–141
Babenko B, Yang M, Belongie S (2011) Robust object tracking with online multiple instance learning. IEEE Trans Pattern Anal Mach Intell 33(8):1619–1632
Yao R, Shi Q, Shen C, Zhang Y, van den Hengel A (2013) Part-based visual tracking with online latent structural learning. In: Proceedings of CVPR
Wu Y, Ma B, Yang M, Zhang J, Jia Y (2014) Metric learning based structural appearance model for robust visual tracking. IEEE Trans Circuits Syst Video Technol 24(5):865–877
Mei X, Ling H (2009) Robust visual tracking using \(\ell 1\) minimization. In: Proceedings of 2009 IEEE 12th international conference on computer vision (ICCV’09), pp 1–8
Kwon J, Lee K (2010) Visual tracking decomposition. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR’10), pp 1269–1276
Yang M, Zhang C, Wu Y, Pei M, Jia Y (2013) Robust object tracking via online multiple instance metric learning. In: Proceedings of multimedia and expo workshops (ICMEW), 2013 IEEE international conference, pp 1–4
Yang M, Wu Y, Pei M, Ma B, Jia Y (2014) Coupling semi-supervised learning and example selection for online object tracking. In: Proceedings of the 12th Asian conference on computer vision (ACCV2014) (in press)
Arif O, Vela A et al. (2009) Non-rigid object localization and segmentation using eigenspace representation. In: Proceedings of IEEE 12th international conference on computer vision (ICCV’09), IEEE, pp 803–808
Porikli F, Tuzel O, Meer P (2006) Covariance tracking using model update based on lie algebra. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR’06), vol 1, pp 728–735
Wu Y, Ma B, Li P (2012) A variational method for contour tracking via covariance matching. Sci China Inf Sci 55(11):2635–2645
Tuzel O, Porikli F, Meer P (2008) Pedestrian detection via classification on riemannian manifolds. IEEE Trans Pattern Anal Mach Intell 30(10):1713–1727
Pang Y, Yuan Y, Li X (2008) Gabor-based region covariance matrices for face recognition. IEEE Trans Circuits Syst Video Technol 18(7):989–993
Arsigny V, Fillard P, Pennec X, Ayache N (2006) Geometric means in a novel vector space structure on symmetric positive-definite matrices. SIAM J Matrix Anal Appl 29(1):328–347
Hartley R (2008) Multiple view geometry in computer vision. Cambridge University Press, Cambridge
Birchfield ST, Rangarajan S (2005) Spatiograms versus histograms for region-based tracking. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR’05), vol 2, pp 1158–1163
Comaniciu D, Ramesh V, Meer P (2003) Kernel-based object tracking. IEEE Trans Pattern Anal Mach Intell 25(5):564–575
Wu Y, Ma B (2014) Learning distance metric for object contour tracking. Pattern Anal Appl 17(2):265–277
Nejhum S, Ho J, Yang M (2010) Online visual tracking with histograms and articulating blocks. Computer Vis Image Underst 114(8):901–914
Zhuang B, Lu H, Xiao Z, Wang D (2014) Visual tracking via discriminative sparse similarity map. Image Process IEEE Trans 23(4):1872–1881
Yang M, Pei M, Wu Y, Ma B, Jia Y (2013) Online-learning structural appearance model for robust visual tracking. In: Proceedings of intelligence science and big data engineering, Springer, New York. pp 30–39
Li X, Hu W, Zhang Z, Zhang X, Zhu M, Cheng J (2008) Visual tracking via incremental log-euclidean riemannian subspace learning. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR’08), vol 0, pp 1–8
Wu Y, Cheng J, Wang J, Lu H (2009) Real-time visual tracking via incremental covariance tensor learning. In: Proceedings of IEEE 12th international conference on computer vision (ICCV’09), pp 1631–1638
Tyagi A, Davis J, Potamianos G (2008) Steepest descent for efficient covariance tracking. In: Proceedings of IEEE workshop on motion and video computing (WMVC’08), IEEE, pp 1–6
Hong X, Chang H, Shan S, Chen X, Gao W (2009) Sigma set: a small second order statistical region descriptor. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR’09), pp 1802–1809
Wu Y, Wu B, Liu J, Lu H (2008) Probabilistic tracking on riemannian manifolds. In: Proceedings of 19th international conference on pattern recognition (ICPR’08), IEEE, pp 1–4
Hu H, Qin J, Lin Y, Xu Y (2008) Region covariance based probabilistic tracking. In: Proceedings of 7th world congress on intelligent control and automation (WCICA’08), pp 575–580
Palaio H, Batista J (2008) Multi-object tracking using an adaptive transition model particle filter with region covariance data association. In: Proceedings of 19th international conference on pattern recognition (ICPR’08), pp 1–4
Hager GD, Belhumeur PN (1998) Efficient region tracking with parametric models of geometry and illumination. Pattern Anal Mach Intell IEEE Trans 20(10):1025–1039
Collins R (2003) Mean-shift blob tracking through scale space. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR’03), vol 2, pp 234–240
Zivkovic Z, Krose B (2004) An em-like algorithm for color-histogram-based object tracking. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR’04), vol 1, pp 798–803
Hu J, Juan C, Wang J (2008) A spatial-color mean-shift object tracking algorithm with scale and orientation estimation. Pattern Recognit Lett 29(16):2165–2173
Zhang H, Huang W, Huang Z, Li L (2005) Affine object tracking with kernel-based spatial-color representation. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR’05), vol 1, pp 293–300
Yilmaz A (2011) Kernel based object tracking using asymmetric kernels with adaptive scale and orientation selection. Mach Vis Appl 22(2):255–268
Tuzel O, Porikli F, Meer P (2008) Learning on lie groups for invariant detection and tracking. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR’08), pp 1–8
Yang F, Lu H, Chen Y (2010) Bag of features tracking. In: Proceedings of pattern recognition (ICPR), 2010 20th international conference on, IEEE, pp 153–156
Acknowledgments
This work was supported in part by the Natural Science Foundation of China (NSFC) under Grant No. 61472036, and the Specialized Fund for Joint Building Program of Beijing Municipal Education Commission.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
In this appendix, we describe the first derivative of \({\varvec{C}}_{R} \big (\psi (\varvec{A})\big )\) with respect to \(\varvec{A}\). Since
we obtain \(\frac{\partial {\varvec{C}}_{R} (\psi (\varvec{A}))}{\partial \varvec{A}}\) as follows. In Eq. (18), \(K^{\prime }(\psi (\varvec{A}))\) denotes the first derivative of Gaussian kernel function \(K(\psi (\varvec{A}))= \exp (-\frac{1}{2} \Vert \psi (\varvec{A})\Vert ^{2}) \) with respect to affine transformation parameters \(\varvec{A}\).
The first derivative of \(\varvec{K(\psi (\varvec{A}))}\) with respect to translation vector \(\varvec{t}\):
The first derivative of \(\varvec{K(\psi (\varvec{A}))}\) with respect to rotation angle \({\varvec{\theta }}\):
where
The first derivative of \(\varvec{K(\psi (\varvec{A}))}\) with respect to skew direction \(\varvec{\phi }\):
where
The first derivative of \(\varvec{K(\psi (\varvec{A}))}\) with respect to scale \(\varvec{\lambda }\):
where
and
Rights and permissions
About this article
Cite this article
Wu, Y., Ma, B. & Jia, Y. Differential tracking with a kernel-based region covariance descriptor. Pattern Anal Applic 18, 45–59 (2015). https://doi.org/10.1007/s10044-014-0430-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10044-014-0430-6