Modeling method for spatial implicit function in large-scale scene

Abstract

The present invention relates to a space for a large scene modeling method of implicit functions. The method comprises the following steps of subjecting the three-dimensional directed point cloud of a large-scale scene to space division and constructing octrees to obtain an octree array, wherein the octree array comprises a series of overlapped octrees; reconstructing a spatial implicit function with gradient constraints and boundary constraints on each octree independently. According to the technical scheme of the invention, the reconstruction parallelism and the reconstruction parallel acceleration of spatial implicit functions are realized, and the reconstruction efficiency of spatial implicit functions is improved. Meanwhile, the surface reconstruction efficiency of large-scale scenes is improved. Moreover, the surface geometric details of reconstructed three-dimensional scenes are maintained, and the realism of the model rendering is enhanced. The reconstruction efficiency is improved and the consumption of computing resources is reduced. In addition, the offline updates of digital urban models might be implemented, and the flexibility and the high efficiency of urban digitization are improved.

Description

technical field [0001] The invention relates to the technical field of three-dimensional model reconstruction, in particular to a spatial implicit function modeling method for large scenes. Background technique [0002] In the field of 3D model reconstruction, spatial implicit function reconstruction technology is a surface reconstruction technology, which is used to convert unstructured 3D directed point clouds into locally smooth spatial implicit functions. The spatial implicit function can reflect the detailed shape of the surface of the 3D directed point cloud. After the triangulation of the spatial implicit function, a 3D grid with a topological structure can be obtained, so as to render the 3D model with high quality. The spatial implicit function has high flexibility and can describe scenes of various complex shapes; due to the use of global constraints, the spatial implicit function is robust to noise and can smoothly handle holes in directed point clouds. [0003] ...

AI Technical Summary

Problems solved by technology

[0004] 1. The spatial implicit function reconstruction method is very time-consuming for large-scale scenes;

[0005] 2. The spatial implicit function reconstruction method tends to over-smooth the detailed structure in the scene

[0007] However, due to the use of global constraints, as the scale of the 3D scene increases, the reconstruction of the spatial implicit function becomes very time-consuming. The hundreds of millions of scaled 3D directed point clouds on ordinary computers usually require several to dozens of hours of computing time. , far from meeting the actual needs

In addition, global constraints will cause the shape details of the scene to be over-smoothed, directly affecting the realistic reproduction of the reconstructed model

Images