Parametric Surface Reconstruction From Point Clouds to Triangular Meshes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing surface reconstruction methods from point clouds face challenges in balancing representation accuracy and computational efficiency, particularly in converting point clouds to triangular meshes, due to the lack of an ordered structure and potential noise, leading to inaccurate reconstructions.
Innovation Solution
A method involving parametric representation using anchor points with spherical harmonic and trigonometric basis functions to convert point clouds into triangular meshes, employing farthest point-sampling and normal estimation algorithms to ensure accurate and efficient surface reconstruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If voxel grid or symbolic distance field is used for surface reconstruction, then representation accuracy is improved, but computational efficiency deteriorates due to dense division of 3D space
Solution Approach 1:
The patent transforms the representation from explicit voxel grid or implicit distance field to parametric surface representation using spherical harmonics. This changes the fundamental parameters of the representation system, allowing high accuracy surface description with significantly reduced computational complexity by representing surfaces through coefficients rather than dense spatial division.
Solution Approach 2:
The patent moves from 3D spatial domain (voxel grid) or function domain (distance field) to a lower-dimensional parameter space using spherical harmonic coefficients. This dimensional transformation enables efficient representation of complex surfaces through a compact set of parameters rather than exhaustive 3D sampling.
2Ease of manufacture
If point clouds data is used directly for surface reconstruction, then data collection ease is improved, but processing difficulty worsens due to lack of ordered structure and topology
Solution Approach 1:
The patent introduces parametric surface representation as an intermediary between raw point clouds and final triangular mesh output. This intermediary layer organizes the unordered point cloud data into a structured parametric form using spherical harmonics, which then can be efficiently converted to triangular meshes, thus reducing the direct processing complexity of raw point clouds.
Solution Approach 2:
The patent segments the surface representation into local patches around anchor points, each represented by spherical harmonic coefficients. This segmentation allows the complex task of processing entire point clouds to be divided into manageable local regions, reducing overall processing complexity while maintaining global accuracy.
3Reliability
If existing surface reconstruction algorithms are used, then basic reconstruction capability is achieved, but representation accuracy and computational efficiency cannot be balanced simultaneously
Solution Approach 1:
The patent combines multiple representation techniques (spherical harmonics for geometric representation, farthest point sampling for point selection, normal estimation for surface orientation) into a composite reconstruction system. This composite approach leverages the strengths of each component to achieve both high accuracy and efficiency, overcoming the limitations of individual methods.
Data Source
AI summary
A method, device, terminal and medium for reconstructing surface from point clouds based on parametric representation are provided. The method includes: acquiring point clouds to be processed corresponding to a target object, and determining a parametric surface of an anchor point based on parametric representation, representing an object surface of the point clouds to be processed by using the parametric surface; converting the parametric surface into a corresponding triangular mesh to reconstruct a 3D surface of the target object; the parametric representation consists of any quantity of anchor points and the anchor points are located inside or outside the target object and face the three-dimensional surface of the target object, and the target object is observed to obtain observation information represented based on positions and orientations of the anchor points and coefficient of a pre-constructed basis function, representing the parametric surface of the anchor points by using the observation information.