3D Model Retrieval Based on Hand Drawn Sketches Using LDA Model

In recent years, hand drawn sketch based 3D model retrieval method has become a popular research area due to the rapid growth of available 3D models on the Internet. However, most existing sketch based 3D model retrieval methods only consider using low-level visual features that cannot capture users'search intention. In this paper, we propose a novel sketch-based 3D model retrieval approach by using LDA method to generate topics address these issues. Topics can provide a more descriptive and discriminative meaningful representation rather than low-level feature descriptors, which span the semantic gap between 3D models and hand-drawn sketches. Firstly, we generate 2D views of 3D models in the database. Then we extract low-level features from query sketch and 2D views of 3D models and encode them as visual words. Lastly, we adopt LDA method to achieve topics distribution based on these visual words. So the 3D models which have similar topics as the query sketch can be returned to the user. The experiment results demonstrate that our method can achieve more accurate retrieval results and have a superior performance than other previously proposed methods which merely used low-level feature descriptors.