Advertisement

A Family of Efficient Appearance Models Based on Histogram of Oriented Gradients (HOG), Color Histogram and Their Fusion for Human Pose Estimation

  • Yong ZhaoEmail author
  • Yong-feng Ju
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 891)

Abstract

Human pose estimation can be addressed within the pictorial structures framework, where a principal difficulty is to model the appearance of body parts. For solving this difficulty, three new models are proposed in this paper. The appearance model based on Histogram of Oriented Gradients (HOG) and Support Vector Data Description (SVDD) is built by the linear combination of sub-classifiers constructed using the SVDD algorithm, while the mixing weights can be learned by using the maximum likelihood estimation algorithm. Moreover, a human part has a specific location probability in different images, then according to learned location probability from static image to be processed, the corresponding color histogram can be calculated, resulting in the appearance model based on color and location probability. According to the illumination and color contrast between clothes and background of the static image to be processed, the respective mixing weights for two appearance models can be learned and then used to build the combined appearance model. We use our appearance models to human pose estimation based on pictorial structure and evaluate them on two image datasets, experiments results show they can get higher pose estimation accuracy.

Keywords

Human pose estimation Appearance model Histogram of Oriented Gradients (HOG) Color histogram 

References

  1. 1.
    Felzenszwalb, P.F., Huttenlocher, D.P.: Pictorial structures for object recognition. Int. J. Comput. Vis. 61(01), 55–79 (2005)CrossRefGoogle Scholar
  2. 2.
    Thomas, B.M., Hilton, A., Krüger, V.: Visual Analysis of Humans, pp. 131–275. Springer, London (2011)Google Scholar
  3. 3.
    Fischler, M., Elschlagr, R.: The representation and matching of pictorial structures. IEEE Trans. Comput. 22(01), 67–92 (1973)CrossRefGoogle Scholar
  4. 4.
    Sapp, B., Toshev, A., Taskar, B.: Cascaded models for articulated pose estimation. In: Proceedings of the 11th European Conference on Computer Vision, pp. 406–420. Springer, Berlin (2010)CrossRefGoogle Scholar
  5. 5.
    Ramanan, D.: Learning to parse images of articulated bodies. In: Proceedings of the 20th Annual Conference on Neural Information Processing Systems, pp. 1129–1136. MIT Press, Cambridge (2006)Google Scholar
  6. 6.
    Andriluka, M., Roth, S., Schiele, B.: Pictorial structures revisited: People detection and articulated pose estimation. In: Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1014–1021. IEEE, Piscataway (2009)Google Scholar
  7. 7.
    Ukita, U.: Articulated pose estimation with parts connectivity using discriminative local oriented contours. In: Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition, pp. 3154–3161. IEEE, Piscataway (2012)Google Scholar
  8. 8.
    Eichner, M., Ferrari, V.: Better appearance models for pictorial structures. In: Proceedings of the 20th British Machine Vision Conference, pp. 3.1–3.11. BMVA Press, Dundee (2009)Google Scholar
  9. 9.
    Tian, T.P., Sclaroff, S.: Fast globally optimal 2d human detection with loopy graph models. In: Proceedings of the 2010 IEEE Conference on Computer Vision and Pattern Recognition, pp. 81–88. IEEE, Piscataway (2010)Google Scholar
  10. 10.
    Sapp, B., Jordan, C., Taskar, B.: Adaptive pose priors for pictorial structures. In: Proceedings of the 2010 IEEE Conference on Computer Vision and Pattern Recognition, pp. 422–429. IEEE, Piscataway (2010)Google Scholar
  11. 11.
    Sun, M., Telaprolu, M., Lee, H., et al.: An efficient branch-and-bound algorithm for optimal human pose estimation. In: Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1616–1623. IEEE, Piscataway (2012)Google Scholar
  12. 12.
    Ferrari, V., Marin-Jimenez, M., Zisserman, A.: Progressive search space reduction for human pose estimation. In: Proceedings of the 2008 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–8. IEEE, Piscataway (2008)Google Scholar
  13. 13.
    Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: Proceedings of the 2005 IEEE Conference on Computer Vision and Pattern Recognition, pp. 886–893. IEEE, Piscataway (2005)Google Scholar
  14. 14.
    Johnson, S., Everingham, M.: Combining discriminative appearance and segmentation cues for articulated human pose estimation. In: Proceedings of the 12th IEEE International Conference on Computer Vision, pp. 405–412. IEEE, Piscataway (2009)Google Scholar
  15. 15.
    Tran, D., Forsyth, D.: Configuration estimates improve pedestrian finding. In: Proceedings of the Twenty-first Annual Conference on Neural Information Processing Systems. MIT Press, Cambridge (2007)Google Scholar
  16. 16.
    Buehler, P., Everingham, M., et al.: Long term arm and hand tracking for continuous sign language TV broadcasts. In: Proceedings of the 19th British Machine Vision Conference, pp. 1105–1114. BMVA Press, UK (2008)Google Scholar
  17. 17.
    Johnson, S.: Articulated Human Pose Estimation in Natural Images [Ph.D. dissertation]. University of Leeds, UK (2012)Google Scholar
  18. 18.
    Dvaid, M.J., Robert, P.W.: Support vector data description. Mach. Learn. 54(1), 45–66 (2004)CrossRefGoogle Scholar
  19. 19.
    Xue, Z.X., Liu, S.Y., Liu, W.L., et al.: SVDD based learning algorithm with progressive transductive support vector machines. Pattern Recog. Artif. Intell. 21(6), 721–727 (2008)Google Scholar
  20. 20.
    Zhu, X.K., Yang, D.G.: Multi-class support vector domain description for pattern recognition based on a measure of expansibility. Acta Electronica Sinica 37(03), 464–469 (2009)Google Scholar
  21. 21.
    Niazmardi, S., Homayouni, S., Safari, S.: An improved FCM algorithm based on the SVDD for unsupervised hyperspectral data classification. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 6(2), 831–839 (2013)CrossRefGoogle Scholar
  22. 22.
    Gurram, P., Kwon, H.: Ensemble learning based on multiple kernel learning for hyperspectral chemical plume detection. In: Proceedings of the SPIE, vol. 7695, pp. 5–9. SPIE Press, Washington (2010)Google Scholar
  23. 23.
    Dvaid, M.J., Robert, P.W.: Support vector domain description. Pattern Recogn. Lett. 20(11–13), 1191–1199 (1999)Google Scholar
  24. 24.
    Jiang, H.: Human pose estimation using consistent max-covering. IEEE Trans. Pattern Anal. Mach. Intell. 33(09), 1911–1918 (2011)CrossRefGoogle Scholar
  25. 25.
    Han, G.J., Zhu, H., Ge, J.R.: Effective search space reduction for human pose estimation with Viterbi recurrence algorithm. Int. J. Model. Identif. Control 18(04), 341–348 (2013)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.School of Electronic and Control EngineeringChang’an UniversityXi’anChina
  2. 2.School of AutomationXi’an University of Posts and TelecommunicationsXi’anChina

Personalised recommendations