CVAIMMOct 11, 2023

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

arXiv:2310.07376v21 citationsh-index: 2
Originality Incremental advance
AI Analysis

This work addresses noise and outlier issues in point cloud data, which is crucial for applications like digital twins and metaverse, but it appears incremental as it builds directly on PointCleanNet.

The paper tackled point cloud denoising and outlier detection by addressing the limitation of PointCleanNet in ignoring local geometric structure, using a Dynamic Graph CNN approach, and reported improved performance with higher AUPR and lower Chamfer Distance compared to conventional methods.

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.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes