Three-dimensional model classification method based on OfficientNet and convolutional neural network

Abstract

The invention discloses a three-dimensional model classification method based on an OfficientNet and a convolutional neural network, and relates to the field of three-dimensional model classification. The problem that an existing three-dimensional model classification method is low in classification precision is solved. The method comprises the following steps: obtaining a two-dimensional view, a shape feature D1, a shape feature D2, a shape feature D3, a Zernike moment feature and a Fourier feature of each side surface of a three-dimensional model; constructing a deep convolutional neural network combining the OfficientNet and the convolutional neural network, fusing the shape feature D1, the shape feature D2, the shape feature D3, the Zernike moment feature and the Fourier feature to form a shape distribution feature vector, and taking the two-dimensional view and the shape distribution feature vector as input of the deep convolutional neural network to obtain a classification result of the three-dimensional model; according to the method, the shape distribution features and the view features are fused, and the precision of three-dimensional model classification is improved.

Description

technical field [0001] The invention relates to the field of three-dimensional model classification, in particular to a three-dimensional model classification method based on EfficientNet and convolutional neural network. Background technique [0002] In recent years, with the increasing application of 3D models, 3D model classification has become an important part of the field of digital geometry. At present, 3D models have been widely used in various fields such as virtual reality, industrial design computer-aided design, cutting-edge unmanned driving, film and television animation, and molecular biology. It has grown exponentially in the Internet. Its complexity and diversity Significant improvement can be seen. How to efficiently classify and efficiently retrieve these 3D models has become the focus of research in related fields, so a lot of research work has emerged. At present, whether it is 3D model classification or 3D model retrieval, how to design a reasonable al...

AI Technical Summary

Problems solved by technology

However, the accuracy of the classification results is low by only using the 2D view image combined with the 3D model obtained by the existing deep learning method

Images