Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Neural Splines: Fitting 3D Surfaces with Infinitely-Wide Neural Networks

About

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.

Francis Williams, Matthew Trager, Joan Bruna, Denis Zorin• 2020

Related benchmarks

TaskDatasetResultRank
Scene ReconstructionSceneNet (test)
Chamfer Distance (CD)0.6
16
Surface ReconstructionDC benchmark 1.0 (test)
Chamfer Distance (GT)0.14
9
Surface ReconstructionShapeNet 260 shapes 15
sCD (mean)5.36e-5
9
Surface ReconstructionGargoyle 1.0 (test)
Chamfer Distance (GT)0.16
9
Surface ReconstructionLord Quas benchmark 1.0 (test)
Chamfer Distance (GT)0.12
9
3D surface reconstructionShapeNet 3k points
Reconstruction Time12.7
9
Surface ReconstructionAnchor 1.0 (test)
Chamfer Distance (GT)0.22
9
Surface ReconstructionDaratech benchmark 1.0 (test)
Chamfer Distance (GT)0.21
9
Showing 8 of 8 rows

Other info

Follow for update