Spherical Convolutional Neural Network for 3D Point Clouds

TL;DR

This paper introduces a spherical convolutional neural network designed for 3D point cloud processing, utilizing spherical kernels and octree partitioning to efficiently capture local geometric structures and improve object classification.

Contribution

It presents a novel neural network architecture that employs spherical convolution kernels and octree-based data structuring for efficient 3D point cloud analysis.

Findings

Effective 3D object classification on benchmark datasets

Spherical kernels capture local geometric features

Octree partitioning exploits data sparsity

Abstract

We propose a neural network for 3D point cloud processing that exploits `spherical' convolution kernels and octree partitioning of space. The proposed metric-based spherical kernels systematically quantize point neighborhoods to identify local geometric structures in data, while maintaining the properties of translation-invariance and asymmetry. The network architecture itself is guided by octree data structuring that takes full advantage of the sparse nature of irregular point clouds. We specify spherical kernels with the help of neurons in each layer that in turn are associated with spatial locations. We exploit this association to avert dynamic kernel generation during network training, that enables efficient learning with high resolution point clouds. We demonstrate the utility of the spherical convolutional neural network for 3D object classification on standard benchmark datasets.