English

Deep Learning Parametrization for B-Spline Curve Approximation

Computational Geometry 2018-07-24 v1 Graphics

Abstract

In this paper we present a method using deep learning to compute parametrizations for B-spline curve approximation. Existing methods consider the computation of parametric values and a knot vector as separate problems. We propose to train interdependent deep neural networks to predict parametric values and knots. We show that it is possible to include B-spline curve approximation directly into the neural network architecture. The resulting parametrizations yield tight approximations and are able to outperform state-of-the-art methods.

Keywords

Cite

@article{arxiv.1807.08304,
  title  = {Deep Learning Parametrization for B-Spline Curve Approximation},
  author = {Pascal Laube and Matthias O. Franz and Georg Umlauf},
  journal= {arXiv preprint arXiv:1807.08304},
  year   = {2018}
}

Comments

Accepted at 3DV 2018

R2 v1 2026-06-23T03:09:56.803Z