Squelette hiérarchique pour la description de formes

Abstract : The skeleton is able to efficiently describe a shape by its global and local geometric properties. Nevertheless, as is, it is difficult to distinguish the branches representing the general appearance of those from details. So, we propose to exploit the natural hierarchy of the skeleton by quantifying the significance of branches. Our method is based on the successive shape smoothings and on the deformability of the skeleton to adapt it to each smoothed shape.
Document type :
Conference papers
Complete list of metadatas

Contributor : Aurélie Leborgne <>
Submitted on : Wednesday, June 22, 2016 - 11:39:02 AM
Last modification on : Tuesday, November 19, 2019 - 2:44:53 AM


  • HAL Id : hal-01335715, version 1


Aurélie Leborgne, Julien Mille, Laure Tougne. Squelette hiérarchique pour la description de formes. RFIA (Reconnaissance de Formes et Intelligence Artificielle), Jun 2016, Clermont-Ferrand, France. ⟨hal-01335715⟩



Record views