Minimum description length with local geometry

Martin Styner, Ipek Oguz, Tobias Heimann, Guido Gerig

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    Establishing optimal correspondence across object populations is essential to statistical shape analysis. Minimizing the description length (MDL) is a popular method for finding correspondence. In this work, we extend the MDL method by incorporating various local curvature metrics. Using local curvature can improve performance by ensuring that corresponding points exhibit similar local geometric characteristics that can't always be captured by mere point locations. We illustrate results on a variety of anatomical structures. The MDL method with a combination of point locations and curvature outperforms all the other methods we analyzed, including traditional MDL and spherical harmonics (SPHARM) correspondence, when the analyzed object population exhibits complex structure. When the objects are of simple nature, however, there's no added benefit to using the local curvature. In our experiments, we did not observe a significant difference in the correspondence quality when different curvature metrics (e.g. principal curvatures, mean curvature, Gaussian curvature) were used.

    Original languageEnglish (US)
    Title of host publication2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Proceedings, ISBI
    Pages1283-1286
    Number of pages4
    DOIs
    StatePublished - 2008
    Event2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI - Paris, France
    Duration: May 14 2008May 17 2008

    Other

    Other2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI
    CountryFrance
    CityParis
    Period5/14/085/17/08

    Fingerprint

    Geometry
    Experiments

    Keywords

    • Correspondence
    • Image registration
    • Image shape analysis
    • Modeling
    • Statistics

    ASJC Scopus subject areas

    • Biomedical Engineering

    Cite this

    Styner, M., Oguz, I., Heimann, T., & Gerig, G. (2008). Minimum description length with local geometry. In 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Proceedings, ISBI (pp. 1283-1286). [4541238] https://doi.org/10.1109/ISBI.2008.4541238

    Minimum description length with local geometry. / Styner, Martin; Oguz, Ipek; Heimann, Tobias; Gerig, Guido.

    2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Proceedings, ISBI. 2008. p. 1283-1286 4541238.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Styner, M, Oguz, I, Heimann, T & Gerig, G 2008, Minimum description length with local geometry. in 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Proceedings, ISBI., 4541238, pp. 1283-1286, 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI, Paris, France, 5/14/08. https://doi.org/10.1109/ISBI.2008.4541238
    Styner M, Oguz I, Heimann T, Gerig G. Minimum description length with local geometry. In 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Proceedings, ISBI. 2008. p. 1283-1286. 4541238 https://doi.org/10.1109/ISBI.2008.4541238
    Styner, Martin ; Oguz, Ipek ; Heimann, Tobias ; Gerig, Guido. / Minimum description length with local geometry. 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Proceedings, ISBI. 2008. pp. 1283-1286
    @inproceedings{6719a467c36f4a1aa86fb86de0c8a80f,
    title = "Minimum description length with local geometry",
    abstract = "Establishing optimal correspondence across object populations is essential to statistical shape analysis. Minimizing the description length (MDL) is a popular method for finding correspondence. In this work, we extend the MDL method by incorporating various local curvature metrics. Using local curvature can improve performance by ensuring that corresponding points exhibit similar local geometric characteristics that can't always be captured by mere point locations. We illustrate results on a variety of anatomical structures. The MDL method with a combination of point locations and curvature outperforms all the other methods we analyzed, including traditional MDL and spherical harmonics (SPHARM) correspondence, when the analyzed object population exhibits complex structure. When the objects are of simple nature, however, there's no added benefit to using the local curvature. In our experiments, we did not observe a significant difference in the correspondence quality when different curvature metrics (e.g. principal curvatures, mean curvature, Gaussian curvature) were used.",
    keywords = "Correspondence, Image registration, Image shape analysis, Modeling, Statistics",
    author = "Martin Styner and Ipek Oguz and Tobias Heimann and Guido Gerig",
    year = "2008",
    doi = "10.1109/ISBI.2008.4541238",
    language = "English (US)",
    isbn = "9781424420032",
    pages = "1283--1286",
    booktitle = "2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Proceedings, ISBI",

    }

    TY - GEN

    T1 - Minimum description length with local geometry

    AU - Styner, Martin

    AU - Oguz, Ipek

    AU - Heimann, Tobias

    AU - Gerig, Guido

    PY - 2008

    Y1 - 2008

    N2 - Establishing optimal correspondence across object populations is essential to statistical shape analysis. Minimizing the description length (MDL) is a popular method for finding correspondence. In this work, we extend the MDL method by incorporating various local curvature metrics. Using local curvature can improve performance by ensuring that corresponding points exhibit similar local geometric characteristics that can't always be captured by mere point locations. We illustrate results on a variety of anatomical structures. The MDL method with a combination of point locations and curvature outperforms all the other methods we analyzed, including traditional MDL and spherical harmonics (SPHARM) correspondence, when the analyzed object population exhibits complex structure. When the objects are of simple nature, however, there's no added benefit to using the local curvature. In our experiments, we did not observe a significant difference in the correspondence quality when different curvature metrics (e.g. principal curvatures, mean curvature, Gaussian curvature) were used.

    AB - Establishing optimal correspondence across object populations is essential to statistical shape analysis. Minimizing the description length (MDL) is a popular method for finding correspondence. In this work, we extend the MDL method by incorporating various local curvature metrics. Using local curvature can improve performance by ensuring that corresponding points exhibit similar local geometric characteristics that can't always be captured by mere point locations. We illustrate results on a variety of anatomical structures. The MDL method with a combination of point locations and curvature outperforms all the other methods we analyzed, including traditional MDL and spherical harmonics (SPHARM) correspondence, when the analyzed object population exhibits complex structure. When the objects are of simple nature, however, there's no added benefit to using the local curvature. In our experiments, we did not observe a significant difference in the correspondence quality when different curvature metrics (e.g. principal curvatures, mean curvature, Gaussian curvature) were used.

    KW - Correspondence

    KW - Image registration

    KW - Image shape analysis

    KW - Modeling

    KW - Statistics

    UR - http://www.scopus.com/inward/record.url?scp=51049098314&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=51049098314&partnerID=8YFLogxK

    U2 - 10.1109/ISBI.2008.4541238

    DO - 10.1109/ISBI.2008.4541238

    M3 - Conference contribution

    SN - 9781424420032

    SP - 1283

    EP - 1286

    BT - 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Proceedings, ISBI

    ER -