3D Object Categorization Based on Histogram of Distance and Normal Vector Angles on Surface Points

This paper propose a method to 3D modelscategorization based on geometric features from face and vertexof any 3D model using probabilistic neural network. For 3Dmodel classification, we use histogram of two variables, i.e., theangle between normal vector on the object surface point andvector that connect shape origin to that point; and distance ofobject surface point to shape origin. Also, for better separabilityof different models, Euclidean distance histogram for pairs ofsurface points is used. The most advantage of using histogram topresent the features is that it leads to reduce the feature vectordimension and consequently computational cost in classificationprocess. Performance of the proposed method is investigatedusing McGill database. The final result shows desiredclassification rate.

Keywords:

component , 3D objects , vertex normal vector , centerto vertex vector , Euclidean distance and histogram