This paper addresses the problem of reconstructing the geometry and color of a Lambertian scene, given some fully calibrated images acquired with wide baselines. In order to completely model the input data, we propose to represent the scene as a set of colored depth maps, one per input image. We formulate the problem as a Bayesian MAP problem which leads to an energy minimization method. Hidden visibility variables are used to deal with occlusion, reflections and outliers. The main contributions of this work are: a prior for the visibility variables that treats the geometric occlusions; and a prior for the multiple depth maps model that smoothes and merges the depth maps while enabling discontinuities. Real world examples showing the efficiency and limitations of the approach are presented.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Bayesian 3D modeling from images using multiple depth maps


    Beteiligte:
    Gargallo, P. (Autor:in) / Sturm, P. (Autor:in)


    Erscheinungsdatum :

    01.01.2005


    Format / Umfang :

    1298683 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Computing depth maps from descent images

    Xiong, Y. / Olson, C. F. / Matthies, L. H. | British Library Online Contents | 2005


    Extracting View-Dependent Depth Maps from a Collection of Images

    Kang, S. B. / Szeliski, R. | British Library Online Contents | 2004



    Refinement of depth maps by fusion of multiple estimates

    Krishnamurthy, B. / Rastogi, A. | British Library Online Contents | 2013


    Sensor calibration using dense depth maps

    KROEGER TILL | Europäisches Patentamt | 2023

    Freier Zugriff