A Kinematics Significance Based Skeleton Map for Rapid Viewpoint Selection

Viewpoint selection has become a very active area of research during the last decade. Computation of good viewpoints is important in graph drawing, scene understanding, etc. A good viewpoint can give us rich information about a scene. In this study, we present a novel Kinematics Significance based Skeleton Map (KSSM) during iterative Laplacian contraction. Inspired by salient viewpoint selection, we propose a new rapid computation method based on a few initial viewpoints for iteration. It allows us to compute the final best viewpoint via Loop subdivision stencil rapidly for iterative decision under a threshold. Experimental results demonstrate that the KSSM can describe the topology information of 3D mesh effectively. It also presents the validity and effectiveness of viewpoint selection based on KSSM.