People Action Recognition in Image Sequences Using a 3D Articulated Object

  • Jean-Charles Atine
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3211)


This article deals with recognition of human action using motion tracking of articulated objects based on its 3D geometric modeling. We minimize a measure of difference between views of a real scene and the corresponding 3D ones. The minimization of the function uses colorimetric information and allows us to automatically estimate the position and the posture of the subject in the scene. We used an hybrid genetic algorithm to optimize the 3D model posture. The matching of views acquired by several video cameras displaced in the real scene and the synthesis views allows inferring the parameters that check the model. Afterward, the movement parameters obtain we will identify the people action using a fuzzy classification.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Quentin Delamarre et Olivier Faugeras, I.N.R.I.A RobotVis Project, 3D articulated Models I.N.R.I.A - Projet Robot Vis (February 2000) Google Scholar
  2. 2.
    Apprentissage artificiel, concepts et algorithmes, ed. EYROLLES Google Scholar
  3. 3.
    Whitley, D.: A genetic algorithm Tutorial, Technical Report CS-93-103, March 10 (1993) Google Scholar
  4. 4.
    Ormoneit, D., Sidenbladh, H., Black, M.J., Hastie, T., Fleet, D.J.: Learning and Tracking Human Motion using Functional Analysis. In: Proc IEEE Workshop Human Modeling, Analysis and Synthesis, June 2000, pp. 2–9 (2000)Google Scholar
  5. 5.
    Ju, S.X., Black, M.J., Yacoob, Y.: Cardboard People : A parametrized Models of Articulated Image Motion Google Scholar
  6. 6.
    Black, M.J., Yacoob, Y.: Tracking and Recognizing Rigid and Non-rigid Facial Motion Using Local Parametric Models of Image Motion. In: ICCV, pp. 374–381 (1995)Google Scholar
  7. 7.
    Kakadiaris, I., Metaxas, D.: Model-Based Estimation of 3D Human Motion. IEEE transactions on patern analysis and machine Intelligence 22(12) (December 2000)Google Scholar
  8. 8.
    Nelder, J.A., Mead, R.: A simplex method for function minimisation. The computer Journal 7, 308–313 (1965)zbMATHGoogle Scholar
  9. 9.
    Perret, Y.: Suivi de parametres de modele géométrique à partir de séquences vidéo multivues , Univeristé Claude Bernard - Lyon1, Laboratoire d’informatique Graphique, Image et modélisation, Décembre 17 (2001) Google Scholar
  10. 10.
    Amat, J.-L. Yahiaoui, G.: Traitement avancées pour le traitement de l’information, Réseaux de neurones, logique floue, algorithmes génétiques Google Scholar
  11. 11.
    Hamada, R., Sakai, S., Tanaka, H.: Scene identification in News video by character region segmentation, National Institute of Informatics Google Scholar
  12. 12.
    Philipp-Foliguet, S., Vieira, M.B.: Segmentation d’images en régions floues, Logique Floue et Applications. In: LFA 2000, La Rochelle (2000)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Jean-Charles Atine
    • 1
  1. 1.INSAToulouse cedex 4France

Personalised recommendations