Surface Reconstruction from Point Cloud Using Skeletal Curves
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for reconstructing surfaces from point clouds, especially those with occlusions and ambiguous topology, fail to produce high-quality results due to limitations in interaction, complexity, and reliability, leading to poor automatic reconstruction and difficulty in handling sharp features.
Innovation Solution
A method and system that extract skeletal curves, edit and assign sweeping paths, fit closed NURBS curves, reconstruct generalized cylinders, and merge them into a single surface, allowing for interactive adjustments to improve surface quality, using techniques like squared distance minimization and curvature-based fitting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If conventional automatic reconstruction methods (Poisson, Radial Basis Function) are used, then the reconstruction process is fully automated, but the reconstruction quality is poor when data is occluded and topology structure is ambiguous
Solution Approach 1:
The patent segments the reconstruction process into distinct modules: automatic skeleton extraction, interactive curve editing, cross-section fitting, and surface generation. This allows the system to combine automated processing with targeted user intervention at critical stages, resolving the contradiction between automation and quality by applying automation where sufficient and human judgment where needed.
Solution Approach 2:
The patent introduces skeletal curves as an intermediary representation between the raw point cloud and the final surface. These curves serve as a mediating structure that captures the essential topology and geometry, allowing users to interact with a simplified representation rather than raw points, thereby improving both automation efficiency and reconstruction quality.
2Ease of manufacture
If surface smoothness assumption is used in reconstruction, then the algorithm is simple to implement, but it is difficult to get good result if the data have sharp features
Solution Approach 1:
The patent applies local quality by allowing different regions of the surface to have different properties. The skeletal curve framework enables sharp features to be explicitly represented where needed, while smooth regions can still benefit from simplified processing. This resolves the contradiction by making the surface representation adaptive to local geometric characteristics rather than uniformly smooth or complex.
3Manufacturing precision
If interactive methods are used to improve reconstruction quality, then user control is enhanced, but the methods are too complicated for users to interact with
Solution Approach 1:
The skeletal curves serve as an intermediary that simplifies user interaction. Instead of requiring users to manipulate raw point clouds or complex surface parameters, they interact with intuitive curve structures that naturally represent the object's topology. This resolves the contradiction by providing user control through a simplified intermediate representation.
Solution Approach 2:
The patent transforms the interaction problem from 3D surface manipulation to 1D curve editing. By working with skeletal curves (1D structures) rather than full surfaces (2D/3D structures), the complexity of interaction is significantly reduced while still enabling control over the final surface reconstruction quality.
4Manufacturing precision
If more interaction is provided to improve reconstruction result, then reconstruction quality improves, but the interaction process becomes more complicated
Solution Approach 1:
The patent segments the interaction process into distinct, manageable stages: skeleton extraction, curve editing, cross-section fitting, and surface generation. Each stage has a specific interaction type and complexity level, allowing users to provide focused input where most needed without overwhelming interaction complexity throughout the entire process.
Data Source
AI summary
A method for reconstructing surface from a point cloud includes following steps: (a) extracting skeletal curves from an input point cloud; (b) editing the extracted skeletal curves, and assigning sweeping path; (c) obtaining sliced point clouds along the edited skeletal curves, and fitting a closed NURBS curve according to the sliced point clouds; (d) reconstructing the point cloud to get generalized cylinders along the assigned sweeping path, according to the closed NURBS curves; (e) merging the generalized cylinders into a single surface, and smoothing intersections of the generalized cylinders so as to reconstruct surface from the point cloud. The invention further relates to a system for reconstructing surface from a point cloud. The invention can reconstruct the surface with high accuracy by the minimum interactions, and can deal with point cloud data having missing region caused by occlusion. In addition, the invention can achieve high reconstruction quality and fine controllability.