Corpus ID: 131774030

DeepSphere: towards an equivariant graph-based spherical CNN

  title={DeepSphere: towards an equivariant graph-based spherical CNN},
  author={Micha{\"e}l Defferrard and Nathanael Perraudin and T. Kacprzak and R. Sgier},
  • Michaël Defferrard, Nathanael Perraudin, +1 author R. Sgier
  • Published 2019
  • Computer Science, Mathematics
  • ArXiv
  • 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… CONTINUE READING
    5 Citations

    Figures and Topics from this paper.

    Explore Further: Topics Discussed in This Paper

    Constant Curvature Graph Convolutional Networks
    • 10
    • PDF
    Functional Penalised Basis Pursuit on Spheres
    • 1
    • PDF


    Learning SO(3) Equivariant Representations with Spherical CNNs
    • 35
    Group Equivariant Convolutional Networks
    • 536
    • PDF
    DeepSphere: Efficient spherical Convolutional Neural Network with HEALPix sampling for cosmological applications
    • 46
    • PDF
    Spherical CNNs
    • 224
    • Highly Influential
    • PDF
    Graph-Based Classification of Omnidirectional Images
    • P. Frossard, R. Khasanova
    • Mathematics, Computer Science
    • 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)
    • 2017
    • 34
    • PDF
    SphereNet: Learning Spherical Representations for Detection and Classification in Omnidirectional Images
    • 90
    • PDF
    Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering
    • 2,339
    • PDF
    Spherical CNNs on Unstructured Grids
    • 52
    • PDF
    On the Generalization of Equivariance and Convolution in Neural Networks to the Action of Compact Groups
    • 153
    • PDF
    Learning Spherical Convolution for Fast Features from 360° Imagery
    • 102
    • PDF