Research on Evaluation of Model Similarity Based on Face Match

Model similarity measurement is an important problem in retrieval of CAD models. In order to compute the similarity degree between two models correctly, a method of model similarity evaluation based on face match is proposed in this paper. Here, face adjacent graph is adopted to describe topological structures in model and construct the face matching matrix between two models. At the same time, greedy algorithm is applied to compute the similarity between these two models. In experiments, the proposed method is used to measure the similarity degree between target model and source model. Experimental results show that the method can evaluate the difference of models efficiently.