Geometric Total Variation for Texture Deformation

Dmitriy Bespalov, Anders Lindbjerg Dahl, Ali Shokoufandeh

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

    340 Downloads (Pure)

    Abstract

    In this work we propose a novel variational method that we intend to use for estimating non-rigid texture deformation. The method is able to capture variation in grayscale images with respect to the geometry of its features. Our experimental evaluations demonstrate that accounting for geometry of features in texture images leads to significant improvements in localization of these features, when textures undergo geometrical transformations. Accurate localization of features in the presense of unkown deformations is a crucial property for texture characterization methods, and we intend to expoit this property for estimation of non-rigid deformations in our future work.
    Original languageEnglish
    Title of host publicationProceedings of 20th International Conference of Pattern Recognition 2010
    Publication date2010
    Pages4597-4600
    ISBN (Print)978-1-4244-7542-1
    Publication statusPublished - 2010
    Event20th International Conference on Pattern Recognition - Istanbul, Turkey
    Duration: 23 Aug 201026 Aug 2010
    Conference number: 20

    Conference

    Conference20th International Conference on Pattern Recognition
    Number20
    Country/TerritoryTurkey
    CityIstanbul
    Period23/08/201026/08/2010

    Fingerprint

    Dive into the research topics of 'Geometric Total Variation for Texture Deformation'. Together they form a unique fingerprint.

    Cite this