A building three-dimensional modeling and virtual scene generation system based on a depth image

Abstract

The invention discloses a building three-dimensional modeling and virtual scene generation method and system based on a depth image. The method comprises the steps of calculating and reconstructing accurate three-dimensional depth data; Optimizing and converting the three-dimensional data of multiple types of sources to obtain a corresponding high-speed streaming point cloud format model, and carrying out point cloud registration; carrying out Projection deformation of the image, and establishing a corresponding relation between adjacent picture pixels; Connecting triangular patches formed bypoint cloud triangularization to generate a building object surface; performing Modeling by using a quadric surface and a complex polygon tool, changing the state of the quadric surface, and processing the irregular polygon to obtain a building three-dimensional model; using Three-dimensional rendering interface software to import the three-dimensional model of the building, and redrawing and displaying the model every preset time, so that the construction of the virtual environment of the building and the generation of the scene under different scenes are realized, and the good sense of reality is achieved.

Description

technical field [0001] The invention relates to the field of three-dimensional modeling, in particular to a system for building three-dimensional modeling and virtual scene generation based on depth images. Background technique [0002] At present, with the rapid development of computer graphics and image processing technology, the 3D virtual scene can reproduce the real scene vividly and vividly from the plane scene pictures, bringing people good visual effects and visual experience. The demand for 3D visualization technology Therefore, how to create the required 3D scene has received more and more attention and research, and has been widely used in various industries. The 3D digital city can establish a virtual urban geographic information environment and provide realistic geographic scene information, which is conducive to the solution of complex problems in urban construction. [0003] In the prior art, it is proposed to use laser radar and aircraft to realize the recon...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, it is proposed to use laser radar and aircraft to realize the reconstruction of 3D real scenes, which has certain applications in open areas and complex environments, but laser scanning equipment is relatively expensive, with a large amount of data, complex operations and relatively high calculations; In addition, for large complex buildings with multiple floors, the aircraft is not easy to control in narrow areas such as corridors, and it is easy to hit a wall and cause damage; it is proposed to use Kinect released by Microsoft to realize the reconstruction of 3D scenes, which has a certain effect on the reconstruction of a single object , but it cannot satisfy the indoor scene rendering of multiple objects; on this basis, it is proposed to use the ICP method to realize the reconstruction of the 3D scene, but due to the limitation of memory, this method also has certain limitations; The standard method realizes 3D scene reconstruction, but there is no precise positioning in the selection of the shooting position, and there are certain defects in the reconstruction effect of the 3D scene

Images