Most dense stereo correspondence algorithms start by establishing discrete pixel matches and later refine these matches to sub-pixel precision. Traditional sub-pixel refinement methods attempt to determine the precise location of points, in the secondary image, that correspond to discrete positions in the reference image. We show that this strategy can lead to a systematic bias associated with the violation of the general symmetry of matching cost functions. This bias produces random or coherent noise in the final reconstruction, but can be avoided by refining both image coordinates simultaneously, in a symmetric way. We demonstrate that the symmetric sub-pixel refinement strategy results in more accurate correspondences by avoiding bias while preserving detail.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Improved sub-pixel stereo correspondences through symmetric refinement


    Contributors:
    Nehab, D. (author) / Rusinkiewiez, S. (author) / Davis, J. (author)


    Publication date :

    2005-01-01


    Size :

    540090 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Improved Sub-pixel Stereo Correspondences through Symmetric Refinement

    Nehab, D. / Rusinkiewicz, S. / Davis, J. et al. | British Library Conference Proceedings | 2005


    Epiflow-A paradigm for tracking stereo correspondences

    Zhang, H. / Negahdaripour, S. | British Library Online Contents | 2008


    An Adaptive Strategy for Finding Stereo Correspondences

    Menard, C. / Kropatsch, W. / Austrian Association for Pattern Recognition | British Library Conference Proceedings | 1997


    Stereo and motion correspondences using nonlinear optimization method

    Park, E. / Wohn, K. | British Library Online Contents | 2006


    A pointwise smooth surface stereo reconstruction algorithm without correspondences

    Brown, R. G. / Chase, J. G. / Hann, C. E. | British Library Online Contents | 2012