Abstract

AbstractIn the field of intelligent manufacturing, how to construct the parts of 3D model has become a major problem in the process of modeling. At present, the major industrial parts manufacturers have developed an available 3D model part library, so an effective search method is needed to help users find the required 3D model from the large 3D model database. The most commonly used 3d model search method is by name, although this method is very accurate, some of the models are only a general name, unable to achieve accurate positioning, so I need another way which has the function of automatic classification search, together with the name of the search method, according to actual needs to choose a suitable method to use. Due to the rapid development of deep learning technology, this paper proposes a method to realize 3D model search by hand sketching. The application of deep learning to the classification and search of 3D models can significantly improve the searching ability of 3D models. In this way, it can realize the search of 3D models more intelligently.KeywordsCNNDiscriminant feature extractionMutual learning strategy

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call