Non-convex hull surface reconstruction via adaptive isosurface extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current surface reconstruction methods from point clouds face challenges in efficiently handling large-scale numerical optimization and producing high-quality polygon meshes, especially with irregular sampling and measurement errors, leading to excessive computational costs and varying reconstruction quality.
Innovation Solution
A computer-implemented method that estimates non-convex hull signed distance function parameters from oriented point data sets and evaluates these functions on volumetric meshes to approximate a smooth, watertight surface using an isosurface algorithm, such as Marching Cubes, allowing for adaptive subsampling and reduced computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If implicit surface reconstruction methods are used to produce watertight surfaces, then surface quality and reliability are improved, but computational cost and device complexity increase due to large-scale numerical optimization problems
Solution Approach 1:
The patent segments the surface reconstruction problem into two independent parts: (1) fitting a signed distance function to oriented points using analytical formulas without numerical optimization, and (2) generating the polygon mesh by evaluating the implicit function on a regular grid and applying isosurface extraction. This segmentation eliminates large-scale numerical optimization while maintaining watertight surface quality.
Solution Approach 2:
The patent performs preliminary action by first fitting the signed distance function parameters analytically before mesh generation. The analytical fitting pre-computes all necessary geometric information, eliminating the need for subsequent numerical optimization during mesh generation, thus reducing overall computational complexity.
2Manufacturing precision
If implicit functions are evaluated on a regular grid of sufficient resolution to produce high-quality meshes, then manufacturing precision is improved, but computational cost and loss of time increase excessively
Solution Approach 1:
The patent applies dynamics by making the grid resolution adaptive rather than uniformly high. The method evaluates the implicit function on a regular grid with resolution adapted to the local features of the surface, using isosurface extraction algorithms that dynamically adjust sampling density. This maintains high mesh quality in critical regions while reducing computational cost in less critical areas.
3Productivity
If adaptive subsampling is applied to reduce data set size, then productivity and computational efficiency are improved, but measurement precision may be compromised
Solution Approach 1:
The patent uses parameter changes by introducing an error threshold parameter that controls the subsampling process. The adaptive subsampling algorithm selectively retains or discards oriented points based on whether their inclusion would exceed the specified error threshold, thus maintaining measurement precision within acceptable bounds while reducing data set size for improved computational efficiency.
Data Source
AI summary
The invention is a computer implemented method, device, system, or article for reconstructing a surface of an object. In particular, the invention comprises estimating a non-convex hull signed distance function parameters from a data set of an object and evaluating the non-convex hull signed distance function on vertices of a volumetric mesh. The invention further comprises approximating the zero level set of the non-convex hull signed distance function by a polygonal mesh using an isosurface algorithm to provide surface reconstruction of an object.