Abstract
A real-time object tracking method based on distribution field (DF) constructs with correlation coefficients is proposed to solve the drawbacks of local search and poor real-time performance exhibited by traditional DF tracking methods. With the goal of adapting to complex environments and changes in tracking speed, we propose an algorithm based on DFs and global searching by dense sampling. First, we use the DFs to construct an appearance model that functions as a target descriptor in the particle filter framework, allowing dynamic updating of the appearance model. Then, we measure the similarity using correlation coefficients based on fast Fourier transforms (FFTs) instead of the L1-norm of DFs to reduce the time complexity, overcome the drawback of randomness when using sparse sampling, and avoid falling into local optima from the gradient descent used in traditional DF methods. The results of experiments show that our proposed algorithm not only performs in real time but is also more robust for a variety of complex environments than those of six state-of-the-art algorithms on eight challenging video sequences.
Similar content being viewed by others
References
Babenko B, Ming-Hsuan Y, Belongie S (2011) Robust object tracking with online multiple instance learning. Patt Anal Mach Intell 33(8):1619–1632
Baker S, Matthews I (2004) Lucas-Kanade 20 years on: a unifying framework. Int J Comput Vis 56(3):221–255
Chenggang Y, Yongdong Z et al (2014) A Highly Parallel framework for HEVC coding unit partitioning tree decision on many-core processors. IEEE Signal Process Lett 21(5):573–576
Chenggang Y, Yongdong Z et al (2014) Parallel deblocking filter for HEVC on many-core processor. Electron Lett 50(5):805–806
Chenggang Y, Yongdong Z, Feng D, Liang L (2013) Highly parallel framework for HEVC motion estimation on many-core platform. Data Compression Conference (DCC), Snowbird, UT, USA
Collins RT (2003) Mean-shift blob tracking through scale space. Comp Vis Patt Recog 2:234–240
Doucet A, de Freitas N, Gordon N (eds) (2001) Sequential Monte Carlo methods in practice. Springer Verlag, New York
Gordon NJ, Salmond DJ, Smith AFM (1993) Novel approach to nonlinear/non-Gaussian Bayesian state estimation. IEE Proceedings 140(2):107–113
Grabner H, Grabner M, Bischof H (2006) Real-time tracking via on-line boosting. BMVC Conference 1:47–56
Grabner H, Leistner C, Bischof H (2008) Semi-supervised on-line boosting for robust tracking. Proc of European Conference on Computer Vision pp 234–247
Hager GD, Dewan M, Stewart CV (2006) Multiple kernel tracking with SSD. Comp Vis Patt Recog 1:798–805
Han-xuan Y, Feng Z, Liang W, Zhan S, Ling S (2011) Recent advances and trends in visual tracking: a review. J Neurocompt 74(18):3823–3831
Heideman MT, Johnson DH, Burrus CS (1984) Gauss and the history of the fast Fourier transform. IEEE ASSP Mag 1(4):14–21
Huang G, Pun C-M, Lin C, Zhou Y (2016) Non-rigid visual object tracking using user-defined marker and Gaussian kernel. Multimed Tools Appl 75(10):5473–5492
Ji-feng N, Lei Z, David Z, Chengke W (2012) Robust mean shift tracking with corrected background-weighted histogram. Comp Vis 6(1):62–69
Kaihua Z, Huihui S (2013) Real-time visual tracking via online weighted multiple instance learning. Pattern Recogn 46(1):397–411
Kai-hua Z, Hui-hui S (2013) Real-time visual tracking via on-line weighted multiple instance learning. Pattern Recogn 46(1):397–411
Kaihua Z, Lei Z, Ming-Hsuan Y (2012) Real-time compressive tracking. Eur Conf Comput Vision 3:864–877
Kai-hua Z, Lei Z, Ming-hsuan Y (2012) Real-time compressive tracking. Proc of European Conference on Computer Vision pp 864–877
Khenouchi H, et al (2016) The Modulus of the Complex Correlation Coefficient Between Co-Polarized Channels for Oil Spill Observation. Living Planet Symposium 740
Kitagawa G (1996) Monte Carlo filter and smoother for non-Gaussian nonlinear state space models. J Comput Graph Stat 5(1):1–25
Koutra D et al (2016) D elta C on: principled massive-graph similarity function with attribution. ACM Transactions on Knowledge Discovery from Data (TKDD) 10.3:28
Lin C, Pun C-M, Huang G (2016) Highly non-rigid video object tracking using segment-based object candidates. Multimedia Tools and Applications 76(7):9565–9586
Liu JS, Chen R (1998) Sequential Monte Carlo methods for dynamic systems. J Am Stat Assoc 93(443):1032–1044
Maggio E, Cavallaro A (2010) Video tracking: Theroy and practice. John Wiley and Sons, London, pp 15–16
Saffari A, Leistner C, Santner J, Godec M, Bischof H (2009) On-line random forests. Proc of International Conference on Computer Vision pp 1393–1400
Sanjeev Arulampalam M, Maskell S, Gordon N, Clapp T (2002) A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking. IEEE Trans Signal Process 50(2):174–188
Sevilla-Lara L, Learned-Miller E (2012) Distribution fields for tracking. Comp Vis Patt Recog 34:1910–1917
Tang A, Scalzo F (2016) Similarity Metric Learning for 2D to 3D Registration of Brain Vasculature. International Symposium on Visual Computing. Springer International Publishing
Viola P, Jones M (2001) Rapid object detection using a boosted Cascade of simple features. Comp Vis Patt Recog 1:511–518
Wang J, et al (2016) Optimizing top precision performance measure of content-based image retrieval by learning similarity function. arXiv preprint arXiv:1604.06620
Yilmaz A, Javed O, Shah M (2006) Object tracking: a survey. Journal of ACM Computing Surveys(CSUR) 38(4):13
Yu X, Xiaojun W, Hongyuan W (2012) Object tracking algorithm based on partial feature combination. Opto-Electron Eng 39(7):67–74
Yuan X, Yan-yun Q, Cui-hua L (2012) Online multiple instance gradient feature selection for robust visual tracking. Pattern Recogn Lett 39(9):1075–1082
Acknowledgment
This work was supported in part by the Research Committee of the University of Macau (MYRG2015-00011-FST, MYRG2015-00012-FST) and the Science and Technology Development Fund of Macau SAR (093-2014-A2).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Qin, P., Pun, CM. Object tracking using distribution fields with correlation coefficients. Multimed Tools Appl 77, 8979–9002 (2018). https://doi.org/10.1007/s11042-017-4790-y
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-017-4790-y