Abstract
AbstractIn this paper we present a novel method for shape comparison of 3D models that employs multilevel spherical moments analysis approach relying on voxelization and spherical mapping of the 3D models. For an input polygon-soup 3D model, firstly a pose normalization step is done to align the model into a canonical coordinate frame and then to define the shape representation with respect to this orientation. Afterward we rasterize its exterior polygons into a cubical voxel grids, then a series of homocentric sphere with their center superposing the center of the voxel grids cut the grids into several spherical surfaces. Finally moments of each sphere are computed and the moments belong to all spheres constitute the descriptor of the model. Experiments show that this shape similarity comparison method outperforms in retrieval performance many previously proposed ones.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
