DeepSphere: towards an equivariant graph-based spherical CNN
2019-04-08Code Available1· sign in to hype
Michaël Defferrard, Nathanaël Perraudin, Tomasz Kacprzak, Raphael Sgier
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/deepsphere/deepsphere-tf1Officialtf★ 15
- github.com/deepsphere/deepsphere-pytorchpytorch★ 165
- github.com/deepsphere/deepsphere-cosmo-tf1tf★ 135
- github.com/deepsphere/deepsphere-weatherpytorch★ 74
- github.com/deepsphere/deepsphere-cosmo-tf2tf★ 23
- github.com/deepsphere/code-iclr20tf★ 0
- github.com/deepsphere/paper-deepsphere-rlgm2019none★ 0
Abstract
Spherical data is found in many applications. By modeling the discretized sphere as a graph, we can accommodate non-uniformly distributed, partial, and changing samplings. Moreover, graph convolutions are computationally more efficient than spherical convolutions. As equivariance is desired to exploit rotational symmetries, we discuss how to approach rotation equivariance using the graph neural network introduced in Defferrard et al. (2016). Experiments show good performance on rotation-invariant learning problems. Code and examples are available at https://github.com/SwissDataScienceCenter/DeepSphere