Log In Sign Up

Towards Efficient Graph Convolutional Networks for Point Cloud Handling

by   Yawei Li, et al.

In this paper, we aim at improving the computational efficiency of graph convolutional networks (GCNs) for learning on point clouds. The basic graph convolution that is typically composed of a K-nearest neighbor (KNN) search and a multilayer perceptron (MLP) is examined. By mathematically analyzing the operations there, two findings to improve the efficiency of GCNs are obtained. (1) The local geometric structure information of 3D representations propagates smoothly across the GCN that relies on KNN search to gather neighborhood features. This motivates the simplification of multiple KNN searches in GCNs. (2) Shuffling the order of graph feature gathering and an MLP leads to equivalent or similar composite operations. Based on those findings, we optimize the computational procedure in GCNs. A series of experiments show that the optimized networks have reduced computational complexity, decreased memory consumption, and accelerated inference speed while maintaining comparable accuracy for learning on point clouds. Code will be available at <>.


page 8

page 14

page 15

page 16

page 17


Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds

We propose a spherical kernel for efficient graph convolution of 3D poin...

A Central Difference Graph Convolutional Operator for Skeleton-Based Action Recognition

This paper proposes a new graph convolutional operator called central di...

Learning to Segment 3D Point Clouds in 2D Image Space

In contrast to the literature where local patterns in 3D point clouds ar...

VA-GCN: A Vector Attention Graph Convolution Network for learning on Point Clouds

Owing to the development of research on local aggregation operators, dra...

Multi-Kernel Diffusion CNNs for Graph-Based Learning on Point Clouds

Graph convolutional networks are a new promising learning approach to de...

SVNet: Where SO(3) Equivariance Meets Binarization on Point Cloud Representation

Efficiency and robustness are increasingly needed for applications on 3D...

Incompleteness of graph convolutional neural networks for points clouds in three dimensions

Graph convolutional neural networks (GCNN) are very popular methods in m...