Variational approximation of a second order free discontinuity problem in computer vision (Q2706229)

From MaRDI portal





scientific article
Language Label Description Also known as
English
Variational approximation of a second order free discontinuity problem in computer vision
scientific article

    Statements

    0 references
    0 references
    0 references
    19 March 2001
    0 references
    theory and algorithms for computer vision
    0 references
    variational problems
    0 references
    free gradient discontinuities
    0 references
    functions of bounded variation
    0 references
    elliptic functionals
    0 references
    Sobolev spaces
    0 references
    De Giorgi \(\Gamma\)-convergence
    0 references
    Variational approximation of a second order free discontinuity problem in computer vision (English)
    0 references
    A version is considered of the functional proposed in [\textit{A. Blake} and \textit{A. Zisserman}, ``Visual Reconstruction'', MIT Press, Cambridge, MA (1987)] for computer vision problems. This functional depends on free discontinuities, free gradient discontinuities, and second order derivatives. It is shown that this functional can be approximated by elliptic functionals defined on Sobolev spaces. Approximation is done in a variational sense (namely, in the sense of the De Giorgi \(\Gamma\)-convergence) and extends an approximation procedure presented in [\textit{L. Ambrosio} and \textit{V. M. Tortorelli}, Commun. Pure Appl. Math. 43, No. 8, 999-1036 (1990; Zbl 0722.49020), and Boll. Unione. Mat. Ital., VII. Ser. B, No. 1, 6, 105-123 (1992; Zbl 0776.49029)]. An application to the problem of computing depth form stereo images is discussed. Some numerical examples are presented.
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references