Multi-resolution multi-object statistical shape models based on the locality assumption

(Medical Image Analysis 2017)

Matthias Wilms, Heinz Handels, and Jan Ehrhardt

Graphical abstract


  • A novel approach for learning statistical shape models from few training samples
  • Combined representation of global and local variability in a single shape model
  • Outperforms other state-of-the-art approaches on a public data set

The published paper can be found here.

A preprint draft version of the published paper can be downloaded here.

Example Matlab code for the proposed approach can be downloaded here.

Slides of our presentation at the german conference "Bildverarbeitung für die Medizin" in Heidelberg, 2017.

Erstellt am 2. Februar 2017 - 15:28 von Ehrhardt. Zuletzt geändert am 15. März 2017 - 18:05 von Ehrhardt.


Susanne Petersen

Neue Rufnummer:
Tel+49 451 3101 5601
Fax+49 451 3101 5604

Gebäude 64 (Informatik)

Ratzeburger Allee 160
23538 Lübeck