Convolutional Networks for Spherical Signals
By Taco Cohen, Mario Geiger, Jonas Köhler, Max Welling. 2017
75 citations recorded by the source corpus.
The success of convolutional networks in learning problems involving planar signals such as images is due to their ability to exploit the translation symmetry of the data distribution through weight sharing. Many areas of science and egineering deal with signals with other symmetries, such as rotation invariant data on the sphere. Examples include climate and weather science, astrophysics, and chemistry. In this paper we present spherical convolutional networks. These networks use convolutions on the sphere and rotation group, which results in rotational weight sharing and rotation equivariance. Using a synthetic spherical MNIST dataset, we show that spherical convolutional networks are very effective at dealing with rotationally invariant classification problems.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Maurice Weiler
- Max Welling
- Kristen Grauman
- Josh Tenenbaum
- Taco Cohen
- Jiajun Wu
- Trevor Darrell
- Li Fei-Fei
In Computer Vision and Pattern Recognition
- Kristen Grauman
- Luc Van Gool
- Jitendra Malik
- Andrea Vedaldi
- Ming-Hsuan Yang
- Trevor Darrell
- Dacheng Tao
- Andrew Zisserman
Citations in the corpus
- Spherical CNNs (2018, background): “A preliminary version of this work appeared as Cohen et al. (2017).”
- Learning to Reconstruct Shapes from Unseen Classes (2018, background): “Recent papers [Cohen et al., 2018, 2017] have studied differentiable, spherical convolution on spherical projections, aiming to preserve rotational equivariance within a neural network.”
- A General Theory of Equivariant CNNs on Homogeneous Spaces (2018, background): “The first use of induced representations and fields in CNNs is [2], and the first CNN on a non-trivial homogeneous space (the Sphere) is [16].”
- Scale-Equivariant Steerable Networks (2019, methodology): “(2018) and rotation equivariant networks in 3D Cohen et al. (2017); Esteves et al. (2018); Cohen et al. (2019). In Freeman & Adelson (1991) authors describe the algorithm for designing steerable filters for rotations. Rotation steerable filters are used in Cohen & Welling (2016b); Weiler et al. (201”
- Learning Spherical Convolution for Fast Features from 360° Imagery (2017, methodology): “Some recent work studies convolution on a sphere [6, 24] using spectral analysis, but those methods require manually annotated spherical images as training data, whereas our method can exploit existing models trained on perspective images as supervision.”