Point Cloud Denoising and Outlier Detection with Local Geometric Structure by Dynamic Graph CNN

TL;DR

This paper introduces a novel point cloud denoising and outlier detection method that leverages local geometric structures using Dynamic Graph CNN, outperforming existing methods in accuracy.

Contribution

It proposes a new approach incorporating dynamic graph convolutional layers to better utilize local geometric information for improved denoising and outlier detection.

Findings

Outperforms conventional methods in outlier detection accuracy (AUPR).

Achieves better denoising results (Chamfer Distance).

Utilizes dynamic graph CNN for local geometric structure modeling.

Abstract

The digitalization of society is rapidly developing toward the realization of the digital twin and metaverse. In particular, point clouds are attracting attention as a media format for 3D space. Point cloud data is contaminated with noise and outliers due to measurement errors. Therefore, denoising and outlier detection are necessary for point cloud processing. Among them, PointCleanNet is an effective method for point cloud denoising and outlier detection. However, it does not consider the local geometric structure of the patch. We solve this problem by applying two types of graph convolutional layer designed based on the Dynamic Graph CNN. Experimental results show that the proposed methods outperform the conventional method in AUPR, which indicates outlier detection accuracy, and Chamfer Distance, which indicates denoising accuracy.