CONTEXT IN ROBOTIC VISION: CONTROL FOR REAL-TIME ADAPTATION

Paolo Lombardi, Virginio Cantoni, Bertrand Zavidovique

Abstract

Nowadays, the computer vision community conducts an effort to produce canny systems able to tackle unconstrained environments. However, the information contained in images is so massive that fast and reliable knowledge extraction is impossible without restricting the range of expected meaningful signals. Inserting a priori knowledge on the operative “context” and adding expectations on object appearances are recognized today as a feasible solution to the problem. This paper attempts to define “context” in robotic vision by introducing a summarizing formalization of previous contributions by multiple authors. Starting from this formalization, we analyze one possible solution to introduce context-dependency in vision: an opportunistic switching strategy that selects the best fitted scenario among a set of pre-compiled configurations. We provide a theoretical framework for “context switching” named Context Commutation, grounded on Bayesian theory. Finally, we describe a sample application of the above ideas to improve video surveillance systems based on background subtraction methods.

References

  1. Coutelle, C., 1995. Conception d'un système à base d'opérateurs de vision rapides, PhD thesis (in French), Université de Paris Sud (Paris 11), Paris, France.
  2. Crowley, J.L., J. Coutaz, G. Rey, P. Reignier, 2002. Perceptual Components for Context Aware Computing. In Proc. UBICOMP2002, Sweden, available at http://citeseer.nj.nec.com/541415.html.
  3. Dessoude, O., 1993. Contrôle Perceptif en milieu hostile: allocation de ressources automatique pour un système multicapteur, PhD thesis (in French), Université de Paris Sud (Paris 11), Paris, France.
  4. Draper, B.A., J.Bins, K.Baek, 1999. ADORE: Adaptive Object Recognition. In Proc. ICVS99, pp. 522-537.
  5. Dubes, R. C., Jain, A. K., 1989. Random Field Models in Image Analysis. In J. Applied Statistics, v. 16, pp. 131-164.
  6. Ebner, M., A. Zell, 1999. Evolving a task specific image operator. In Proc. 1st European Wshops on Evolutionary Image Analysis, Signal Processing and Telecommunications, Göteborg, Sweden, SpringerVerlag, pp. 74-89.
  7. Horswill, I., 1995. Analysis of Adaptation and Environment. In Artificial Intelligence, v.73(1-2), pp. 1-30, 1995.
  8. Kittler, J., J. Matas, M. Bober, L. Nguyen, 1995. Image interpretation: Exploiting multiple cues. In Proc. Int. Conf. Image Processing and Applications, Edinburgh, UK, pp. 1-5.
  9. Kruppa, H., M. Spengler, B. Schiele, 2001. Context-driven Model Switching for Visual Tracking. In Proc. 9th Int. Symp. Intell. Robotics Sys., Toulouse, France.
  10. Lombardi, P., 2003. A Model of Adaptive Vision System: Application to Pedestrian Detection by Autonomous Vehicles. PhD thesis (in English), Università di Pavia (Italy) and Université de Paris XI (France).
  11. Merlo, X., 1988. Techniques probabilistes d'intégration et de contrôle de la perception en vue de son exploitation par le système de décision d'un robot, PhD thesis (in French), Université de Paris Sud (Paris 11), Paris, France.
  12. Rabiner, L.R., 1989. A tutorial on hidden Markov models. In Proceedings of the IEEE, vol. 77, pp. 257-286.
  13. Rimey, R.D., 1993. Control of Selective Perception using Bayes Nets and Decision Theory. Available at http:// citeseer.nj.nec.com/rimey93control.html.
  14. Roli, F., G. Giacinto, S.B. Serpico, 2001. Classifier Fusion for Multisensor Image Recognition. In Image and Signal Processing for Remote Sensing VI, Sebastiano B. Serpico, Editor, Proceedings of SPIE, v. 4170, pp.103-110.
  15. Rosenfeld, A., R.A. Hummel, S.W. Zucker, 1976. Scene labeling by relaxation operations. In IEEE Trans. Syst. Man Cybern., v. 6, pp. 420-433.
  16. Shekhar, C., S. Kuttikkad, R. Chellappa, 1996. KnowledgeBased Integration of IU Algorithms. In Proc. Image Understanding Workshop, ARPA, v. 2, pp. 1525-1532, 1996.
  17. Stauffer, C., W.E.L. Grimson, 1999. Adaptive Background Mixture Models for Real-Time Tracking. In Proc. IEEE Conf. Comp. Vis. Patt. Rec. CVPR99, pp. 246- 252.
  18. Strat, T.M., 1993. Employing Contextual Information in Computer Vision. In Proc. DARPA93, pp. 217-229.
  19. Tissainayagam, P., D. Suter, 2003. Contour tracking with automatic motion model switching. In Pattern Recognition.
  20. Torralba, A., K.P. Murphy, W.T. Freeman, M.A. Rubin, 2003. Context-based vision system for place and object recognition. In Proc. ICCV'03, available at http://citeseer.nj.nec.com/torralba03contextbased.html.
  21. Toyama, K., E.Horvitz, 2000. Bayesian Modality Fusion: Probabilistic Integration of Multiple Vision Algorithms for Head Tracking. In Proc. ACCV'00, 4th Asian Conf. Comp. Vision, Tapei, Taiwan, 2000.
Download


Paper Citation


in Harvard Style

Lombardi P., Cantoni V. and Zavidovique B. (2004). CONTEXT IN ROBOTIC VISION: CONTROL FOR REAL-TIME ADAPTATION . In Proceedings of the First International Conference on Informatics in Control, Automation and Robotics - Volume 3: ICINCO, ISBN 972-8865-12-0, pages 135-142. DOI: 10.5220/0001143601350142


in Bibtex Style

@conference{icinco04,
author={Paolo Lombardi and Virginio Cantoni and Bertrand Zavidovique},
title={CONTEXT IN ROBOTIC VISION: CONTROL FOR REAL-TIME ADAPTATION},
booktitle={Proceedings of the First International Conference on Informatics in Control, Automation and Robotics - Volume 3: ICINCO,},
year={2004},
pages={135-142},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001143601350142},
isbn={972-8865-12-0},
}


in EndNote Style

TY - CONF
JO - Proceedings of the First International Conference on Informatics in Control, Automation and Robotics - Volume 3: ICINCO,
TI - CONTEXT IN ROBOTIC VISION: CONTROL FOR REAL-TIME ADAPTATION
SN - 972-8865-12-0
AU - Lombardi P.
AU - Cantoni V.
AU - Zavidovique B.
PY - 2004
SP - 135
EP - 142
DO - 10.5220/0001143601350142