Bootstrap-based normal reconstruction

  • Authors:
  • Ahmad Ramli;Ioannis Ivrissimtzis

  • Affiliations:
  • School of Engineering and Computing Sciences, Durham University, UK;School of Engineering and Computing Sciences, Durham University, UK

  • Venue:
  • Proceedings of the 7th international conference on Curves and Surfaces
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose a bootstrap-based method for normal estimation on an unorganised point set. Experimental results show that the accuracy of the method is comparable with the accuracy of the widely used Principal Component Analysis. The main advantage of our approach is that the variance of the normals over the bootstrap samples can be used as a confidence value for the estimated normal. In a proposed application, we use the confidence values to construct a bilateral Gaussian filter for normal smoothing.