Since its original publication, the Semi-Global Matching (SGM) technique has been re-implemented by many researchers and companies. The method offers a very good trade off between runtime and accuracy, especially at object borders and fine structures. It is also robust against radiometric differences and not sensitive to the choice of parameters. Therefore, it is well suited for solving practical problems. The applications reach from remote sensing, like deriving digital surface models from aerial and satellite images, to robotics and driver assistance systems. This paper motivates and explains the method, shows current developments as well as examples from various applications.


    Access

    Download


    Export, share and cite



    Title :

    Semi-Global Matching - Motivation, Developments and Applications


    Contributors:

    Conference:

    2011 ; Stuttgart, Germany


    Published in:

    Publication date :

    2011-09-01


    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    Half-Resolution Semi-Global Stereo Matching

    Hermann, S. / Morales, S. / Klette, R. et al. | British Library Conference Proceedings | 2011


    Real-time stereo vision: Optimizing Semi-Global Matching

    Michael, Matthias / Salmen, Jan / Stallkamp, Johannes et al. | IEEE | 2013


    LARGE SCALE SEMI-GLOBAL MATCHING ON THE CPU

    Spangenberg, R. / Langner, T. / Adfeldt, S. et al. | British Library Conference Proceedings | 2014


    REAL-TIME STEREO VISION: OPTIMIZING SEMI-GLOBAL MATCHING

    Michael, M. / Salmen, J. / Stallkamp, J. et al. | British Library Conference Proceedings | 2013


    Stereo Processing by Semi-Global Matching and Mutual Information

    Hirschmüller, Heiko | German Aerospace Center (DLR) | 2008

    Free access