Scalable and Equivariant Spherical CNNs by Discrete-Continuous (DISCO) Convolutions.
Jeremy OcampoMatthew A. PriceJason D. McEwenPublished in: CoRR (2022)
Keyphrases
- discrete space
- continuous domains
- continuous action
- discrete variables
- continuous space
- continuous variables
- continuous data
- continuous query processing
- discrete data
- continuous functions
- cellular neural networks
- lightweight
- discrete fourier transform
- three dimensional
- piecewise linear
- rotation invariant
- finite number
- spherical harmonics
- continuous state spaces
- continuous valued
- discrete version
- discrete valued
- data streams
- vector valued
- highly scalable
- euclidean space
- fourier transform
- single image
- genetic algorithm