Volumetric Convolution: Automatic Representation Learning in Unit Ball

Abstract

Convolution is an efficient technique to obtain abstract feature representations using hierarchical layers in deep networks. Although performing convolution in Euclidean geometries is fairly straightforward, its extension to other topological spaces---such as a sphere (${\mathbb{S}}^2$) or a unit ball (${\mathbb{B}}^3$)---entails unique challenges. In this work, we propose a novel `\emph{volumetric convolution}' operation that can effectively convolve arbitrary functions in ${\mathbb{B}}^3$. We develop a theoretical framework for \emph{volumetric convolution} based on Zernike polynomials and efficiently implement it as a differentiable and an easily pluggable layer for deep networks. Furthermore, our formulation leads to derivation of a novel formula to measure the symmetry of a function in ${\mathbb{B}}^3$ around an arbitrary axis, that is useful in 3D shape analysis tasks. We demonstrate the efficacy of proposed volumetric convolution operation on a possible use-case i.e., 3D object recognition task.

References in corpus (5)

• Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
• Generative and Discriminative Voxel Modeling with Convolutional Neural Networks
• 3D Steerable CNNs: Learning Rotationally Equivariant Features in Volumetric Data
• SO-Net: Self-Organizing Network for Point Cloud Analysis
• RotationNet: Joint Object Categorization and Pose Estimation Using Multiviews from Unsupervised Viewpoints

Cited by in corpus (1)

• Conditional Generative Modeling via Learning the Latent Space