On the Implementation of Centerline Extraction Based on Confidence Vote in Accumulation Map

Bertrand Kerautret (1) , Adrien Krähenbühl (1) , Isabelle Debled-Rennesson (1) , Jacques-Olivier Lachaud (2)

1 ADAGIO - Applying Discrete Algorithms to Genomics and Imagery
2 LAMA - Laboratoire de Mathématiques

This paper focuses on the implementation details of a recent method which extracts the centerline of 3D shapes using solely partial mesh scans of these shapes. This method [9] extracts the shape centerline by constructing an accumulation map from input points and normal vectors and by filtering it with a confidence vote. This paper presents in details all the algorithms of the method and describes the implementation and development choices. Some experiments test the robustness to the parameter variability and show the current limitations allowing to consider further improvements.

Domaines: Traitement des images [eess.IV]; Vision par ordinateur et reconnaissance de formes [cs.CV]

Bertrand Kerautret, Adrien Krähenbühl, Isabelle Debled-Rennesson, Jacques-Olivier Lachaud. On the Implementation of Centerline Extraction Based on Confidence Vote in Accumulation Map. RRPR 2016 1st Workshop on Reproducible Research in Pattern Recognition, Bertrand Kerautret; Miguel Colom; Pascal Monasse, Dec 2016, Cancùn, Mexico. pp.116-130, ⟨10.1007/978-3-319-56414-2_9⟩. ⟨hal-01571434⟩