3D model retrieval is an important part of multimedia retrieval. However, the performance of traditional retrieval method of text-based and content-based is not satisfying because the keyword and the shape feature do not include the semantic information of the 3D models. The paper explores an ontology and SWRL-based 3D model retrieval system Onto3D. It can infer 3D models' semantic property by rule engine and retrieve the target models by ontology. The experiments on Princeton Shape Benchmark show that Onto3D achieves good performance not only in text retrieval but also in shape retrieval. © 2008 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Wang, X., Lv, T., Wang, S., & Wang, Z. (2008). An ontology and SWRL based 3D model retrieval system. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4993 LNCS, pp. 335–344). https://doi.org/10.1007/978-3-540-68636-1_32
Mendeley helps you to discover research relevant for your work.