In this work, we propose a model for video scenes that contains temporal variability in shape and appearance. We propose a conditionally linear model akin to a dynamic extension of active appearance models. We formulate the problem variationally, and propose a framework where a model complexity cost dictates the "modeling responsibility" of each of the factors: appearance, shape and motion. We render the learning problem well-posed by reverting to a physical and a dynamic prior, and use the finite element method to compute a numerical solution. We illustrate our model to learn and simulate the shape, appearance, and motion of scenes that exhibit some form of temporal regularity, intended in a statistical sense.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Modeling dynamic scenes with active appearance


    Contributors:
    Doretto, G. (author)


    Publication date :

    2005-01-01


    Size :

    325161 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Background Modeling and Subtraction of Dynamic Scenes

    Monnet, A. / Mittal, A. / Paragios, N. et al. | British Library Conference Proceedings | 2003


    Background modeling and subtraction of dynamic scenes

    Monnet, / Mittal, / Paragios, et al. | IEEE | 2003


    Coarse-to-fine adaptive masks for appearance matching of occluded scenes

    Edwards, J. L. / Murase, H. | British Library Online Contents | 1998


    Robust text detection in natural scenes using text geometry and visual appearance

    Yan, S. Y. / Xu, X. X. / Liu, Q. S. | British Library Online Contents | 2014


    Active appearance models with occlusion

    Gross, R. / Matthews, I. / Baker, S. | British Library Online Contents | 2006