Your browser does not support JavaScript!
http://iet.metastore.ingenta.com
1887

access icon free Online learning of mixture experts for real-time tracking

  • Author(s): S. Gu 1 ; Z. Ma 1 ; M. Xie 2 ; Z. Chen 3
    • View affiliations
    • Affiliations: 1: School of Communication and Information Engineering, University of Electronic Science and Technology of China, Chengdu, People's Republic of China ;
      2: School of Electronic Engineering, University of Electronic Science and Technology of China, Chengdu, People's Republic of China ;
      3: Faculty of Computer Science, Memorial University of Newfoundland, St. John's, Canada
  • Source: Volume 10, Issue 6, September 2016, p. 585 – 592
    DOI:  10.1049/iet-cvi.2015.0210 , Print ISSN 1751-9632, Online ISSN 1751-9640
© The Institution of Engineering and Technology
Received 12/06/2015, Accepted 20/02/2016, Revised 16/01/2016, Published 02/03/2016

Template tracking has been extensively investigated in computer vision to track objects for various applications. Tracking based on gradient descent algorithm using image gradient is one of the most popular object tracking method. However, it is difficult to define the relationship between the observed data set and the warping function due to the unobserved heterogeneity of the data set which inevitably results in poor tracking performance. This study proposes a novel method based on hierarchical mixture of expert to perform robust, real-time tracking from stationary cameras. By extending the idea of hyperplane approximation, the proposed approach establishes a hierarchical mixture of generalised linear regression model instead of a single model which reduces the non-linear error. The experiments’ results show significant improvement over the traditional hyperplane approximation (HA) approach.

Inspec keywords: object tracking; image sensors; mixture models; computer vision; learning (artificial intelligence); regression analysis; gradient methods; approximation theory

Other keywords: object tracking method; observed data set; gradient descent algorithm; hyperplane approximation; online mixture experts learning; template tracking; generalised linear regression model; stationary cameras; computer vision; warping function

Subjects: Other topics in statistics; Image sensors; Other topics in statistics; Optical, image and video signal processing; Knowledge engineering techniques; Interpolation and function approximation (numerical analysis); Computer vision and image processing techniques; Interpolation and function approximation (numerical analysis)

References

    1. 1)
      • 17. Ripley, B.D.: ‘Pattern recognition and neural networks’ (The press syndicate of the University of Cambridge, 1996).
    2. 2)
      • 19. Jordan, M.I.: ‘Hierarchical mixtures of experts and the em algorithm’, IEE Colloquium on Advances in Neural Networks for Control and Systems , 1994, 6, pp. 1/1–1/3.
    3. 3)
      • 14. Holzer, S., Ilic, S., Navab, N.: ‘Adaptive linear predictors for real-time tracking’. 2010 IEEE Conf. on Computer Vision and Pattern Recognition (CVPR). IEEE, 2010, pp. 1807–1814.
    4. 4)
      • 18. Grn, B., Leisch, F.: ‘Finite mixtures of generalized linear regression models’. Recent Advances in Linear Models and Related Areas, Physica-Verlag HD, 2008, pp. 205–230.
    5. 5)
      • 8. Holzer, S., Ilic, S., Tan, D., et al: ‘Efficient learning of linear predictors for template tracking’, Int. J. Comput. Vis., 2015, 111, (1), pp. 12–28.
    6. 6)
      • 10. Xiong, X., De la Torre, F.: ‘Global supervised descent method’. Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition, 2015, pp. 2664–2673.
    7. 7)
      • 20. Dumortier, F.A.G.: ‘Theory and practice of recursive identification: Lennart ljung and torsten sderstrm’, Automatica, 1985, 21, (4), pp. 499–501.
    8. 8)
      • 21. Baker, S., Matthews, I.: ‘Lucas-kanade 20 years on: A unifying framework’, Int. J. Comput. Vis., 2004, 56, (3), pp. 221–255.
    9. 9)
      • 4. Grabner, H., Bischof, H.: ‘On-line boosting and vision’.  2006 IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, June 2006, vol. 1, pp. 260–267.
    10. 10)
      • 6. Holzer, S., Pollefeys, M., Ilic, S., et al: ‘Online learning of linear predictors for real-time tracking’. InComputer Vision ECCV 2012, Springer Berlin Heidelberg, 2012, (volume 7572 of Lecture Notes in Computer Science), pp. 470–483.
    11. 11)
      • 2. Babenko, B., Yang, M.-H., Belongie, S.: ‘Visual tracking with online multiple instance learning’. IEEE Conf. on Computer Vision and Pattern Recognition, 2009. CVPR 2009.June 2009, pp. 983–990.
    12. 12)
      • 3. Kalal, Z., Matas, J., Mikolajczyk, K.: ‘Online learning of robust object detectors during unstable tracking’.  2009 IEEE 12th Int. Conf. on Computer Vision Workshops (ICCV Workshops), September 2009, pp. 1417–1424.
    13. 13)
      • 7. Choi, H.-C., Oh, S.-Y.: ‘Robust segment-based object tracking using generalized hyperplane approximation’, Pattern Recogn., 2012, 45, (8), pp. 2980–2991.
    14. 14)
      • 16. Bishop, C.M.: ‘Pattern recognition and machine learning’ (springer, New York, 2006), vol. 1.
    15. 15)
      • 5. Jurie, F., Dhome, M.: ‘Hyperplane approximation for template matching’, IEEE Trans. Pattern Anal. Mach. Intell., 2002, 24, (7), pp. 996–1000.
    16. 16)
      • 1. Collins, R.T., Liu, Y., Leordeanu, M.: ‘Online selection of discriminative tracking features’, IEEE Trans. Pattern Anal. Mach. Intell., 2005, 27, (10), pp. 1631–1643.
    17. 17)
      • 11. Raja, Y., McKenna, S.J., Gong, S.: ‘Segmentation and tracking using colour mixture models’. In Computer Vision ACCV'98, Springer Berlin Heidelberg, 1997, (vol. 1351 of Lecture Notes in Computer Science), pp. 607–614.
    18. 18)
      • 15. Xiong, X., De la Torre, F.: ‘Supervised descent method and its applications to face alignment’.  2013 IEEE Conf. on Computer Vision and Pattern Recognition (CVPR). IEEE, 2013, pp. 532–539.
    19. 19)
      • 12. McKenna, S.J., Raja, Y., Gong, S.: ‘Tracking colour objects using adaptive mixture models’, Image Vis. Comput., 1999, 17, (34), pp. 225–231.
    20. 20)
      • 9. Benhimane, S., Malis, E.: ‘Homography-based 2d visual servoing’. Proc. 2006 IEEE Int. Conf. on Robotics and Automation, 2006. ICRA 2006.May 2006, pp. 2397–2402.
    21. 21)
      • 13. Hager, G.D., Belhumeur, P.N.: ‘Efficient region tracking with parametric models of geometry and illumination’, IEEE Trans. Pattern Anal. Mach. Intell., 1998, 20, (10), pp. 1025–1039.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-cvi.2015.0210
Loading

Related content

content/journals/10.1049/iet-cvi.2015.0210
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading
Print this page
This is a required field
Please enter a valid email address