Item request has been placed! ×
Item request cannot be made. ×
loading  Processing Request

Is affine invariance well defined on SPD matrices? A principled continuum of metrics

Item request has been placed! ×
Item request cannot be made. ×
loading   Processing Request
  • Additional Information
    • Contributors:
      COMUE Université Côte d'Azur (2015-2019) (COMUE UCA); E-Patient : Images, données & mOdèles pour la médeciNe numériquE (EPIONE); Centre Inria d'Université Côte d'Azur; Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria); ERC G-Statistics No 786854ANR UCAJEDI No ANR-15-IDEX-01; G-Statistics; Inria@SiliconValley; GeomStats; ANR-15-IDEX-0001,UCA JEDI,Idex UCA JEDI(2015); European Project: 786854,ERC-2017-ADG,ERC-2017-ADG,G-Statistics(2018)
    • Publication Information:
      CCSD
      Springer
    • Publication Date:
      2019
    • Collection:
      HAL Université Côte d'Azur
    • Subject Terms:
    • Abstract:
      International audience ; Symmetric Positive Definite (SPD) matrices have been widelyused in medical data analysis and a number of different Riemannian met-rics were proposed to compute with them. However, there are very fewmethodological principles guiding the choice of one particular metric fora given application. Invariance under the action of the affinetransfor-mations was suggested as a principle. Another concept is based on sym-metries. However, the affine-invariant metric and the recently proposedpolar-affine metric are both invariant and symmetric. Comparing thesetwo cousin metrics leads us to introduce much wider families: power-affineand deformed-affine metrics. Within this continuum, we investigate otherprinciples to restrict the family size.
    • Relation:
      info:eu-repo/semantics/altIdentifier/arxiv/1906.01349; info:eu-repo/grantAgreement//786854/EU/Foundations of Geometric Statistics and Their Application in the Life Sciences/G-Statistics; ARXIV: 1906.01349
    • Accession Number:
      10.1007/978-3-030-26980-7_52
    • Online Access:
      https://hal.science/hal-02147020
      https://hal.science/hal-02147020v1/document
      https://hal.science/hal-02147020v1/file/main.pdf
      https://doi.org/10.1007/978-3-030-26980-7_52
    • Rights:
      info:eu-repo/semantics/OpenAccess
    • Accession Number:
      edsbas.432D1AC8