This project implements a deep learning pipeline for graph classification using SplineConv in torch_geometric with MPI Faust dataset. This geometric data with spatial features is an appropriate application of SplineConv for constructing a CNN, meaning each kernel has a fully connected set of trainable weight. The algorithm defining the geometric convolution with B-Spline kernels can be found at (Fey, Lenssen, Weichert, M ̈uller (2017) "SplineCNN: Fast Geometric Deep Learning with Continuous B-Spline Kernels".
Fey, Matthias, et al. "Splinecnn: Fast geometric deep learning with continuous b-spline kernels." Proceedings of the IEEE conference on computer vision and pattern recognition. 2018.
Bogo, Federica, et al. "FAUST: Dataset and evaluation for 3D mesh registration." Proceedings of the IEEE conference on computer vision and pattern recognition. 2014.