Research of 3D Model for Information Retrieval Methods Based on Semantic Tree

Abstract

The paper borrowed researching achievements in text retrieval field, describe the semantics of single model by using annotation, and present the semantics of 3D models by using semantic tree, based on the semantic similarity between the retrieval terms and semantic tree nodes with WordNet, and return the model that has strong semantic correlation. A flexible return policy is proposed to filter the representative model of the semantic relevant nodes so that the users can further optimize the retrieval results. According to the experimental results, the method of 3D model retrieval method based-on semantic tree proposed in this paper can improve the efficiency of information retrieval, which is valuable in theory as well as in application.