Stroke type interaction method for real-time partition of three-dimensional gridding model

Abstract

The invention relates to a stroke type interacting method for dividing a 3 d mesh model in real time. The method comprises: executing dividing operations to the 3 d mesh model through an audio-visual stoke interactive interface by users, firstly, using the characteristic perception scale on the basis of angles to obtain the discrete differential geometry characteristics, adopting a diffusion algorithm to carry out the dividing operations to the 3 d mesh model, and then using two rapid boundary adjusting algorisms with unchangeable topology and unchangeable geometry to generate a natural gridding dividing boundary on the basis of convex hull characteristics and the apparent angle characteristics respectively. Users can easily divide the 3D model without much time or technique. The invention provides two characteristic perception scales to divide models, the scales have clear physical meaning and can reflect the discrete differential geometry characteristics, and furthermore two rapid boundary adjusting algorisms with unchangeable topology and unchangeable geometry are respectively provided on the basis of the convex hull characteristics and the apparent angle characteristics. Compared with a dividing method of the prior art, the method can easily realize multi-stage division in one interaction.

Description

technical field [0001] The invention relates to the technical field of computer graphics, and proposes a stroke-style interactive framework for real-time segmentation of 3D models. Background technique [0002] In the field of computer graphics, mesh model segmentation is a very useful tool that can be used for modeling, texture mapping, intersection parameterization, shape gradient, shape retrieval and collision detection, etc. Currently, triangular meshes are a common representation of 3D data because of their expressiveness and speed. However, the triangular mesh itself only characterizes the low-level data structure and cannot reflect high-level structures such as semantic features. In order to obtain a meaningful object segmentation, it is critical to extract semantic features from these underlying data representations. However, the notion of "meaningful" segmentation is very context-dependent and varies depending on the user's subjective wishes. Therefore, it is sti...

AI Technical Summary

Problems solved by technology

However, in these automatic segmentation methods, in order to adapt to different scenarios, some control parameters still need to be adjusted and set by the user

Therefore, due to the ill-posed nature of the problem itself, it is still difficult to achieve automatic segmentation in the strict sense

Images