Abstract: A huge scattered point data includes all kinds of scanning artifacts including noise,outliers,holes and irregular/anisotropic sampling. Most common surface reconstruction methods fail due to these shortcomings. This poses challenges to recovering datasets topology and retrieving features. In order to solve this problem, a robust and efficient reconstruction method was proposed. First, computed local properties for each data point, then used this information to detect simple primitive shapes in the data, at last, described a novel method to extract and optimize boundary curves on the primitive shapes and Employed the reconstructed boundary curves to extract a piccewise smooth surface mesh. The experimental results show the effectiveness of our method with reconstructions of synthetic datasets and real-scenes datasets.

Key words: Scattered point data, Primitive shapes, Boundary curves, Feature points, Multi-resolution analysis, Statistics optumzation