SplineCNN: Fast Geometric Deep Learning with Continuous B-Spline Kernels

CVPR 2018 Matthias FeyJan Eric LenssenFrank WeichertHeinrich Müller

We present Spline-based Convolutional Neural Networks (SplineCNNs), a variant of deep neural networks for irregular structured and geometric input, e.g., graphs or meshes. Our main contribution is a novel convolution operator based on B-splines, that makes the computation time independent from the kernel size due to the local support property of the B-spline basis functions... (read more)

PDF Abstract

Evaluation Results from the Paper


 SOTA for Node Classification on Cora (using extra training data)

     Get a GitHub badge
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK USES EXTRA
TRAINING DATA
COMPARE
Node Classification Cora SplineCNN Accuracy 89.48% # 1