Advertisement

Egomotion Estimation as an Appearance-Based Classification Problem

  • Pedro Sánchez
  • Cornelio Yáñez
  • Jonathan Pecero
  • Apolinar Ramírez
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4225)

Abstract

In this paper a probabilistic approach is considered to develop a methodology to solve the problem of estimation of the position of the observer. The base of this methodology is the appearance vision with which an environment map is constructed using Kernel PCA. For the experiments an image set is acquired in unknown locations in the same environment. The performance of Kernel PCA technique was tested according to the optimum dimension of the environment model and the quantity of images correctly classified using a Bayesian algorithm. To validate the results obtained with Kernel PCA the same experiments were performed with PCA and APEX techniques, then the results were compared showing that Kernel PCA has better performance than PCA and APEX.

Keywords

Egomotion estimation probabilistic approach Kernel PCA 

References

  1. 1.
    Longuet-Higgins, H.C., Prazdny, K.: The Interpretation of a Moving Retinal Image. In: Proc. of the Royal Society B, London (1980)Google Scholar
  2. 2.
    Lourakis, M.I.A.: Using Constraint Lines for Estimating Egomotion. In: Proc. of ACCV 2000, Taipei, Taiwan, January 2000, vol. 2, pp. 971–976 (2000)Google Scholar
  3. 3.
    Convertino, G., Branca, A., Distante, A.: Focus of Expansion Estimation with a Neural Network. In: IEEE International Conference on Neural Networks, June 1996, pp. 1693–1697 (1996)Google Scholar
  4. 4.
    Tomasi, C., Kanade, T.: Factoring images sequences into shape and motion. In: Proc. IEEE Workshop o Visual Motion, pp. 21–28 (1991)Google Scholar
  5. 5.
    Negadharipour, S.: Direct Passive Navigation. PhD thesis, Department of Mechanical Engineering, MIT (1986)Google Scholar
  6. 6.
    Daniilidis, K., Spetsakis, M.: Understanding noise sensitivity in structure from motion. In: Aloimonos, Y. (ed.) Visual Navigation, pp. 61–88 (1996)Google Scholar
  7. 7.
    Burger, W., Bhanu, B.: Estimating 3D egomotion from perspective image sequence. Pattern Analysis and Machine Intelligence, IEEE, Honeywell Syst. & Res. Center, Minneapolis, MN, November 1990, pp. 1040–1058 (1990)Google Scholar
  8. 8.
    Murase, H., Nayar, S.K.: Visual learning and recognition of 3D objects from appearance. International Journal of Computer Vision 14(1), 5–24 (1995)CrossRefGoogle Scholar
  9. 9.
    Cai, Q., Aggarwal, J.K.: Tracking Human Motion in Structured Environments Using Distributed-Camera System. IEEE Transactions on Pattern Analysis, 1241–1247 (1999)Google Scholar
  10. 10.
    Turk, M., Pentland, A.: Eigenfaces for recognition. J. Cog. Neurosci. 3(1), 71–86 (1991)CrossRefGoogle Scholar
  11. 11.
    Krose, B.J., Bunschoten, A.R.: Probabilistic Location by Appearance Models and Active Vision. In: Proceedings of the IEEE International Conference on Robotics and Automation, pp. 2255–2260 (1999)Google Scholar
  12. 12.
    Yang, M.H.: Kernel eigenfaces vs. kernel fisherfaces: Face recognition using kernel methods. In: Proc. IEEE International Conference on Automatic Face and Gesture Recognition, pp. 215–220 (2002)Google Scholar
  13. 13.
    Pourraz, F., Crowley, J.: Continuity properties of the appearance manifold for mobile robot position estimation. In: Proceedings of the 2nd IEEE Workshop on Perception for Mobile Agents, Ft. Collins, CO, pp. 741–752. IEEE Press, Los Alamitos (2001)Google Scholar
  14. 14.
    Muller, K., Mika, S., Ratsch, G., Tsuda, K., Scholkopf, B.: An introduction to kernel-based learning algorithms. IEEE Transactions on Neural Networks, 181–201 (March 2001)Google Scholar
  15. 15.
    Schölkopf, B., Smola, A.J., Müller, K.R.: Nonlinear component analysis as a kernel eigenvalue problem. Neural Comput. 10, 1299–1319 (1998)CrossRefGoogle Scholar
  16. 16.
    Krose, B.J.A., Bunschoten, R.: Probabilistic Location by Appearance Models and Active Vision. In: Proceedings of the IEEE International Conference on Robotics and Automation, pp. 2255–2260 (1999)Google Scholar
  17. 17.
    Diamantaras, K.I., Kung, Y.: A Neural Network Learning Algorithm for Adaptive Principal Components Extraction, pp. 860–864. Princeton UniversityGoogle Scholar
  18. 18.
    Oja, E.: A simplified Neuron Model as a Principal Componet Analyser. J. Math. Biology 15, 267–273Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Pedro Sánchez
    • 1
  • Cornelio Yáñez
    • 2
  • Jonathan Pecero
    • 3
  • Apolinar Ramírez
    • 1
  1. 1.Instituto Tecnológico de Ciudad MaderoTamaulipasMéxico
  2. 2.Centro de Investigación en ComputaciónInstituto Politécnico NacionalMéxico D.F.
  3. 3.Instituto Nacional Politécnico de GrenobleFrance

Personalised recommendations