A comparative study of existing metrics for 3D-mesh segmentation evaluation

Authors:
Halim Benhabiles;Jean-Philippe Vandeborre;Guillaume Lavoué;Mohamed Daoudi
Affiliations:
University of Lille, LIFL (UMR USTL/CNRS 8022), Lille, France;University of Lille, LIFL (UMR USTL/CNRS 8022), Lille, France and TELECOM Lille 1, Institut TELECOM, Lille, France;Université de Lyon, CNRS, INSA-Lyon, LIRIS, UMR5205, 69621, Lyon, France;University of Lille, LIFL (UMR USTL/CNRS 8022), Lille, France and TELECOM Lille 1, Institut TELECOM, Lille, France
Venue:
The Visual Computer: International Journal of Computer Graphics
Year:
2010

Citing 0
Cited 6

SMI 2011: Full Paper: A comparative evaluation of foreground/background sketch-based mesh segmentation algorithms

Computers and Graphics
Sketch-based mesh cutting: A comparative study

Graphical Models
A graph-based technique for semi-supervised segmentation of 3D surfaces

Pattern Recognition Letters
SHREC'12 track: 3D mesh segmentation

EG 3DOR'12 Proceedings of the 5th Eurographics conference on 3D Object Retrieval
SMI 2013: New evaluation metrics for mesh segmentation

Computers and Graphics
On 3D object retrieval benchmarking

3D Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we present an extensive experimental comparison of existing similarity metrics addressing the quality assessment problem of mesh segmentation. We introduce a new metric, named the 3D Normalized Probabilistic Rand Index (3D-NPRI), which outperforms the others in terms of properties and discriminative power. This comparative study includes a subjective experiment with human observers and is based on a corpus of manually segmented models. This corpus is an improved version of our previous one (Benhabiles et al. in IEEE International Conference on Shape Modeling and Application (SMI), 2009). It is composed of a set of 3D-mesh models grouped in different classes associated with several manual ground-truth segmentations. Finally the 3D-NPRI is applied to evaluate six recent segmentation algorithms using our corpus and the Chen et al.’s (ACM Trans. Graph. (SIGGRAPH), 28(3), 2009) corpus.