Visualizing Shape Deformations with Variation of Geometric Spectrum

摘要: This paper presents a novel approach based on spectral geometry to quantify and visualize non-isometric deformations of 3D surfaces by mapping two manifolds. The proposed method can determine multi-scale, through the variation Laplace-Beltrami spectrum shapes. Given triangle meshes, spectra be varied from one another with scale function defined each vertex. is expressed as linear interpolation eigenvalues In iteration step, quadratic programming problem constructed, our derived theorem smoothness energy constraint, compute variation. derivation solution such problem. Therefore, final solved integral which, in turn, quantitatively describes between To evaluate method, we conduct extensive experiments synthetic real data. We employ epilepsy patient imaging data shape left right hippocampi epileptic brains. addition, use longitudinal Alzheimer compare deformation diseased healthy hippocampus. order show accuracy effectiveness also it spatial registration-based methods, e.g., non-rigid Iterative Closest Point (ICP) voxel-based method. These demonstrate advantages