Sharp feature preserving MLS surface reconstruction based on local feature line approximations

Abstract

. Sharp feature preserving MLS surface reconstruction based on local feature line approximations. Graphical Models, Elsevier, 2012, 74 (6) AbstractSharp features in manufactured and designed objects require particular attention when reconstructing surfaces from unorganized scan point sets using moving least squares (MLS) fitting. It's an inherent property of MLS fitting that sharp features are smoothed out. Instead of searching for appropriate new fitting functions our approach computes a modified local point neighborhood so that a standard MLS fitting can be applied enhanced by sharp features reconstruction.We present a two-stage algorithm. In a pre-processing step sharp feature points are marked first. This algorithm is robust to noise since it is based on Gauss map clustering. In the main phase, the selected feature points are used to locally approximate the feature curve and to segment and enhance the local point neighborhood. The MLS projection thus leads to a piecewise smooth surface preserving all sharp features. The method is simple to implement and able to preserve line-type features as well as corner-type features during reconstruction.

Outlier detection and robust normal-curvature estimation in mobile laser scanning 3D point cloud data

“…Methods are proposed and tailored according to their suitability for the particular application e.g. plane fitting [29,30], surface reconstruction [3,31], sharp feature preserving [6,16] and normal estimation [11,32,33].…”

“…Reliable surface reconstruction, object modelling and rendering applications heavily depend on how well the estimated local surface normals and curvatures approximate the true normals and curvatures of the scanned surface [1,2]. Many studies on accurate normal and curvature estimation have been carried out over the years in computer graphics, computer vision, pattern recognition, photogrammetry, reverse engineering, remote sensing and robotics [2][3][4][5][6][7].…”

Segmentation-based Multi-scale Edge Extraction to Measure the Persistence of Features in Unorganized Point Clouds

“…In this line, several authors (Weber et al, 2010;Weber et al, 2012;Gumhold et al, 2001;Feng et al, 2014) propose a region growing method that decomposes the point cloud into clusters, and identifies the regions with sharp features based on the analysis of the normals. In these approaches, extracting sharp edge features from a 3D point cloud requires the computation of accurate normals from the neighborhood to generate high quality surfaces.…”

Voronoi-based feature curves extraction for sampled singular surfaces

“…Recently, Weber et al [6] apply Gauss map clustering [7] to find feature points and locally fit feature curves as cubic Beizer splines. Dey et al [1] use Gaussianweighted graph Laplacian to identify singular points and Reeb graph to connect them into feature curves.…”