Multimedia Tools and Applications

, Volume 72, Issue 1, pp 905–924 | Cite as

2D scale-adaptive tracking based on projective geometry

  • Zhongyu Lou
  • Guang Jiang
  • Chengke Wu


Object tracking is a fundamental challenge in computer vision. For real-time tracking, the efficiency and robustness of the Mean-shift algorithm makes it a popular choice. However, the scale of the Mean-shift kernel is a crucial parameter and no clear mechanism exists presently for updating the scale when a size-changing object is tracked. In this paper, a new method is introduced using projective geometry to determine the size of the object, and in turn the scale of the Mean-shift kernel. In the initial step of the algorithm, the geometric information of the scene is obtained automatically (or manually). With the geometric information, the size of the object is updated. The experimental results show that this algorithm is stable, efficient and outperforms the Mean-shift baseline and other kernel updating methods, such as CAMSHIFT.


Mean-shift Projective geometry Tracking Scale adaptation 


  1. 1.
    Benfold B, Reid I (2011) Unsupervised learning of a scene-specific coarse gaze estimator. In: ICCVGoogle Scholar
  2. 2.
    Bradski GR (1998) Computer vision face tracking for use in a perceptual user interface. In: IEEE workshop on applications of computer vision, Princeton, NJ, pp 214–219Google Scholar
  3. 3.
    Canny J (1986) A computational approach to edge detection. IEEE Trans Pattern Anal Mach Intell 8(6):679–698CrossRefGoogle Scholar
  4. 4.
    Cheng Y (1995) Mean shift, mode seeking, and clustering. IEEE Trans Pattern Anal Mach Intell 17(8):790–799CrossRefGoogle Scholar
  5. 5.
    Chu DM, Smeulders AWM (2010) Thirteen hard cases in visual tracking. In: IEEE workshop on performance evaluation of tracking and surveillanceGoogle Scholar
  6. 6.
    Collins RT (2003) Mean-shift blob tracking through scale space. In: IEEE Computer Society conference on computer vision and pattern recognition, Vancouver, vol II, pp 234–240Google Scholar
  7. 7.
    Comaniciu D, Ramesh V, Meer P (2000) Real-time tracking of non-rigid objects using mean shift. In: IEEE international conference on computer vision and pattern recognition, South Carolina, pp 142–149Google Scholar
  8. 8.
    Comaniciu D, Ramesh V, Meer P (2001) The variable bandwidth mean shift and data-driven scale selection. In: International conference on computer vision, vol I, pp 438–445Google Scholar
  9. 9.
    Comaniciu D, Ramesh V, Meer P (2003) Kernel-based object tracking. IEEE Trans Pattern Anal Mach Intell 25(5):564–577CrossRefGoogle Scholar
  10. 10.
    Everts I, van Gemert JC, Gevers Th (2012) Per-patch descriptor selection using surface and scene properties. In: European conference on computer visionGoogle Scholar
  11. 11.
    Hartley R, Zisserman A (2003) Multiple view geometry in computer vision, 2nd edn. Cambridge University Press, CambridgeGoogle Scholar
  12. 12.
    Hedau V, Hoiem D, Forsyth D (2009) Recovering the spatial layout of cluttered room. In: Proc. ICCVGoogle Scholar
  13. 13.
    Lindeberg T (1998) Feature detection with automatic scale selection. Int J Comput Vis 30(2):79–116CrossRefGoogle Scholar
  14. 14.
    Lou Z, Jiang G, Wu C (2009) Mean-shift tracking of variable kernel based on projective geometry. In: International congress on image and signal processing, TianjinGoogle Scholar
  15. 15.
    Lou Z, Jiang G, Jia L, Wu C (2010) Monocular 3D tracking of mean-shift with scale adaptation based on projective geometry. In: The international conference on multimedia technology, NingboGoogle Scholar
  16. 16.
    Lucena M, Fuertes JM, Blanca NP, Jimenez MJM (2010) Tracking people in video sequences using multiple models. Multimedia Tools Appl 49(2):371–403CrossRefGoogle Scholar
  17. 17.
    Peng N, Yang J, Liu Z, Zhang F (2005) Automatic selection of kernel-bandwidth for mean-shift object tracking. J Softw 16(9):1542–1550CrossRefGoogle Scholar
  18. 18.
    Qi SM, Huang XW, Yi HF (2007) Object tracking by anisotropic kernel mean shift. J Electron Inf Technol 29(3):686–689Google Scholar
  19. 19.
    Sicre R, Nicolas H (2011) Improved Gaussian mixture model for the task of object tracking. In: 14th international conference on computer analysis of images and patternsGoogle Scholar
  20. 20.
    Sotelo MA, Rodriguez FJ, Magdalena L, Bergasa LM, Boquete L (2004) A color vision-based lane tracking system for autonomous driving on unmarked roads. Auton Robots 16(1):95–116CrossRefGoogle Scholar
  21. 21.
    Tyagi A, Keck M, Davis JW, Potamianos G (2006) Kernel-based 3D tracking. In: IEEE conference on computer vision and pattern recognitionGoogle Scholar
  22. 22.
    Yilmaz A (2007) Object tracking by asymmetric kernel mean shift with automatic scale and orientation selection. In: IEEE conference on computer vision and pattern recognition. IEEE, Minneapolis, pp 1–6Google Scholar
  23. 23.
    Zhou H, Yuan Y, Shi CM (2009) Object tracking using SIFT features and mean shift. Comput Vis Image Understand 113:345–352CrossRefGoogle Scholar
  24. 24.
    Zhu S (2006) A study of mean shift and correlative algorithm in visual tracking. PhD thesis, ZheJiang UniversityGoogle Scholar
  25. 25.
    Zhu S, Zhu S (2006) Algorithm of target tracking based on mean shift with adaptive bandwidth of kernel function. J Optoelectron Eng 33(8):11–16Google Scholar
  26. 26.
    Zivkovic Z, Krose B (2004) An EM-like algorithm for color-histogram-based object tracking. In: Proc. IEEE conference on computer vision and pattern recognition, vol 1, pp 798–803Google Scholar
  27. 27.
    Zivkovic Z, Krose B (2004) A probabilistic model for an EM-like object tracking algorithm using color histograms. In: Proc. 6th IEEE int. workshop on performance evaluation of tracking and surveillanceGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  1. 1.State Key Lab. of Integrated Service Networks, School of Telecommunications EngineeringXidian UniversityXianChina

Personalised recommendations