Method for searching for 3D model with 2D pictures

Abstract

The invention relates to the field of data search and provides a cross-model method for searching for a 3D model with 2D pictures. The method comprises following steps: 1) establishing a 3D model library; 2) extracting characteristic vectors; 3) training convolutional neural networks; 4) inputting the characteristic vectors; 5) completing matching retrieval. According to the method, 3D objects and 2D pictures are projected into a new space for measuring the similarity of 3D and 2D; therefore, the problem that the similarity of 3D and 2D data cannot be measured and retrieved due to different data formats is solved; at the same time, the invention also provides an end to end solution scheme which is more efficient than other traditional frames and has better real-time performance.

Description

technical field [0001] The invention relates to the field of data search, in particular to a cross-model method for searching 3D models by using 2D pictures. Background technique [0002] 3D digital models have been widely used in our daily life, such as 3D printing, computer numerical control manufacturing (CNM), 3D film and television, virtual reality (VR), finite element simulation analysis (FEA) and other fields. In life, there are also many other applications, for example, building a 3D scene model based on a 2D image. [0003] Among the existing methods for building 3D scene models based on 2D images, the most traditional method is to perform 3D modeling for each object separately, and then build a 3D module in a given scene. However, 3D modeling often takes a lot of time, especially for the situation where a complete 3D scene needs to be established, and it is difficult to complete the entire subject in real time. Another approach is to search for 3D models based on...

AI Technical Summary

Problems solved by technology

However, it is very difficult to directly match different data types, so this method is not easy to implement

Images