Mesh Segmentation with the Geodesic Means Clustering of Sharp Vertices

K.-H. Yoo, Y.-J. Park, C. Park, W. Li, and J.-S. Ha (Korea)


Mesh segmentation, sharp vertex, geodesic distance, and k -means clustering


In this paper, we adapt the k -means clustering technique to segmenting a given 3D mesh. In order to avoid the locally minimal convergence and speed up the computing time, we extract sharp vertices from the mesh by analyzing its curvature and convexity that respectively reflect the local and global geometry from the viewpoint of cognitive science. The sharp vertices are partitioned into k clusters by iterated converging with the k -means clustering method based on the geodesic distance between each pair of the sharp vertices. For obtaining the effective result of k -means clustering method, we automatically compute a reasonable number of clusters as an initial value k .

Important Links:

Go Back