Three-dimensional model search method based on topology and visual feature

Abstract

The invention discloses a three-dimensional model search method based on topology and visual feature. The method comprises the following steps of: inputting a three-dimensional model to be searched; carrying out standardization preprocessing on the input three-dimensional model, that is to say, transforming the three-dimensional model to a uniform coordinate system through translation transformation, scaling transformation and rotated transformation; constructing a topology structure for the transformed three-dimensional model by a multilayer Reeb picture method; rendering an image of each key topology point of the topology structure; extracting the visual feature of the rendered three-dimensional model; computing similarity according to the extracted visual feature point and the three-dimensional model in a model feature library; and returning the searched result to the user according to the similarity size. According to the three-dimensional model search method provided by the invention, a rigid body model and a non-rigid body model are searched quickly and accurately.

Description

technical field [0001] The invention relates to the field of image retrieval, in particular to a three-dimensional model retrieval method based on topology and visual features. Background technique [0002] With the development of 3D data acquisition technology, 3D graphics modeling method and graphics hardware technology, more and more 3D model databases of real objects have been produced, such as 3D games and industrial entity databases. 3D model is the fourth type of multimedia data after sound, image and video. The increasingly developed Internet and computers with enhanced functions provide conditions for people to share and process 3D models. For the current trillions of 3D models, as well as a large number of 3D models generated every day, how to effectively manage and retrieve has become an urgent problem to be solved. The 3D model retrieval method is the same as other multimedia data retrieval methods, mainly divided into text-based retrieval and content-based retr...

AI Technical Summary

Problems solved by technology

The disadvantage is that the ability of these features to describe the similarity between models is generally not strong enough, and the content description of the 3D model is not sufficient

The disadvantage is: due to the need for different conditions in the process of function projection, it is easy to lose some important information representing the three-dimensional structure

The disadvantage is that when the topology changes, it may have a great impact on the matching structure

Images