Neural Splines: Fitting 3D Surfaces with Infinitely-Wide Neural Networks
Authors:
- 1 June 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 9944-9953
Abstract
We present Neural Splines, a technique for 3D surface reconstruction that is based on random feature kernels arising from infinitely-wide shallow ReLU networks. Our method achieves state-of-the-art results, outperforming recent neural network-based techniques and widely used Poisson Surface Reconstruction (which, as we demonstrate, can also be viewed as a type of kernel method). Because our approach is based on a simple kernel formulation, it is easy to analyze and can be accelerated by general techniques designed for kernel-based learning. We provide explicit analytical expressions for our kernel and argue that our formulation can be seen as a generalization of cubic spline interpolation to higher dimensions. In particular, the RKHS norm associated with Neural Splines biases toward smooth interpolants.Keywords
All Related Versions
Funding Information
- Alfred P. Sloan Foundation
- Samsung
This publication has 22 references indexed in Scilit:
- A Survey of Surface Reconstruction from Point CloudsComputer Graphics Forum, 2016
- Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification2015
- Screened poisson surface reconstructionACM Transactions on Graphics, 2013
- A benchmark for surface reconstructionACM Transactions on Graphics, 2013
- Fast Poisson disk sampling in arbitrary dimensions2007
- Sparse surface reconstruction with adaptive partition of unity and radial basis functionsGraphical Models, 2006
- Reconstruction and representation of 3D objects with radial basis functions2001
- Surface reconstruction from unorganized points1992
- A fast algorithm for particle simulationsJournal of Computational Physics, 1987
- Asymptotic formulas for the dual radon transform and applicationsMathematische Zeitschrift, 1987