DeepAI
Log In Sign Up

PDO-eS^2CNNs: Partial Differential Operator Based Equivariant Spherical CNNs

04/08/2021
by   Zhengyang Shen, et al.
0

Spherical signals exist in many applications, e.g., planetary data, LiDAR scans and digitalization of 3D objects, calling for models that can process spherical data effectively. It does not perform well when simply projecting spherical data into the 2D plane and then using planar convolution neural networks (CNNs), because of the distortion from projection and ineffective translation equivariance. Actually, good principles of designing spherical CNNs are avoiding distortions and converting the shift equivariance property in planar CNNs to rotation equivariance in the spherical domain. In this work, we use partial differential operators (PDOs) to design a spherical equivariant CNN, PDO-eS^2CNN, which is exactly rotation equivariant in the continuous domain. We then discretize PDO-eS^2CNNs, and analyze the equivariance error resulted from discretization. This is the first time that the equivariance error is theoretically analyzed in the spherical domain. In experiments, PDO-eS^2CNNs show greater parameter efficiency and outperform other spherical CNNs significantly on several tasks.

READ FULL TEXT

page 1

page 2

page 3

page 4

10/12/2020

Spherical Convolutional Neural Networks: Stability to Perturbations in SO(3)

Spherical signals are useful mathematical models for data arising in man...
01/30/2018

Spherical CNNs

Convolutional Neural Networks (CNNs) have become the method of choice fo...
03/31/2018

Snap Angle Prediction for 360^∘ Panorama

360^∘ panoramas are a rich medium, yet notoriously difficult to visualiz...
01/07/2019

Spherical CNNs on Unstructured Grids

We present an efficient convolution kernel for Convolutional Neural Netw...
08/07/2022

PDO-s3DCNNs: Partial Differential Operator Based Steerable 3D CNNs

Steerable models can provide very general and flexible equivariance by f...
02/10/2022

Spherical Transformer

Using convolutional neural networks for 360images can induce sub-optimal...
02/11/2019

Gauge Equivariant Convolutional Networks and the Icosahedral CNN

The idea of equivariance to symmetry transformations provides one of the...