We present a monocular object tracker, able to detect and track multiple objects in non-controlled environments. Bayesian per-pixel classification is used to build a tracking framework that segments an image into foreground and background objects, based on observations of object appearances and motions. Gaussian mixtures are used to build the color appearance models. The system adapts to changing lighting conditions, handles occlusions, and works in real-time.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Bayesian Pixel Classification for Human Tracking


    Contributors:
    Roth, Daniel (author) / Doubek, Petr (author) / Gool, Luc Van (author)


    Publication date :

    2005-01-01


    Size :

    464978 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Sub-Pixel Bayesian Estimation of Albedo and Height

    Shekarforoush, H. / Berthod, M. / Zerubia, J. et al. | British Library Online Contents | 1996


    Classification of areas using pixel-by-pixel and sample classifiers

    Dejesusparada, N. / Kumar, R. / Niero, M. et al. | NTRS | 1979


    Bayesian Classification Scheme

    Stutz, John / Cheeseman, Peter / Taylor, Will et al. | NTRS | 1992


    Gibbs likelihoods for Bayesian tracking

    Roth, S. / Sigal, L. / Black, M.J. | IEEE | 2004


    Multistatic Bayesian extended target tracking

    Vivone, Gemine / Braca, Paolo / Granstrom, Karl et al. | IEEE | 2016