Spherical CNNs
This addresses the demand for models that can analyze spherical images in domains such as robotics and weather modeling, representing a novel method for a known bottleneck.
The authors tackled the problem of applying convolutional neural networks to spherical images, which is needed for applications like omnidirectional vision and climate modeling, by introducing spherical CNNs with a rotation-equivariant cross-correlation definition and efficient computation via a generalized FFT, demonstrating effectiveness in 3D model recognition and atomization energy regression.
Convolutional Neural Networks (CNNs) have become the method of choice for learning problems involving 2D planar images. However, a number of problems of recent interest have created a demand for models that can analyze spherical images. Examples include omnidirectional vision for drones, robots, and autonomous cars, molecular regression problems, and global weather and climate modelling. A naive application of convolutional networks to a planar projection of the spherical signal is destined to fail, because the space-varying distortions introduced by such a projection will make translational weight sharing ineffective. In this paper we introduce the building blocks for constructing spherical CNNs. We propose a definition for the spherical cross-correlation that is both expressive and rotation-equivariant. The spherical correlation satisfies a generalized Fourier theorem, which allows us to compute it efficiently using a generalized (non-commutative) Fast Fourier Transform (FFT) algorithm. We demonstrate the computational efficiency, numerical accuracy, and effectiveness of spherical CNNs applied to 3D model recognition and atomization energy regression.