Using surface variability characteristics for segmentation of deformable 3D objects with application to piecewise statistical deformable model

摘要

To cope with the small sample size problem in the construction of Statistical Deformable Models (SDM), this paper proposes two novel measures that quantify the similarity of the variability characteristics among deform-ing 3D meshes. These measures are used as the basis of our proposed technique for partitioning a 3D mesh for the con-struction of piecewise SDM in a divide-and-conquer strat-egy. Specifically, the surface variability information is ex-tracted by performing a global principal component analy-sis on the set of sample meshes. An iterative face clustering algorithm is developed for segmenting a mesh that favors grouping triangular faces having similar variability charac-teristics into a same mesh component. We apply the pro-posed mesh segmentation algorithm to the construction of piecewise SDM and evaluate the representational ability of the resulting piecewise SDM through the reconstruction of unseen meshes. Experimental results show that our approach outperforms several state-of-the-art methods in terms of the representational ability of the resulting piecewise SDM as evaluated by the reconstruction accuracy.