Graph Smoothing for Enhanced Local Geometry Learning in Point Cloud Analysis

TL;DR

This paper introduces a graph smoothing technique combined with local geometry learning to improve the analysis of 3D point clouds, addressing issues of sparse and noisy graph connections.

Contribution

It proposes a novel graph smoothing module that enhances graph structure and local geometry features for better point cloud analysis.

Findings

Improved accuracy in point cloud classification.

Enhanced segmentation performance on real-world datasets.

Robustness to noisy and sparse graph connections.

Abstract

Graph-based methods have proven to be effective in capturing relationships among points for 3D point cloud analysis. However, these methods often suffer from suboptimal graph structures, particularly due to sparse connections at boundary points and noisy connections in junction areas. To address these challenges, we propose a novel method that integrates a graph smoothing module with an enhanced local geometry learning module. Specifically, we identify the limitations of conventional graph structures, particularly in handling boundary points and junction areas. In response, we introduce a graph smoothing module designed to optimize the graph structure and minimize the negative impact of unreliable sparse and noisy connections. Based on the optimized graph structure, we improve the feature extract function with local geometry information. These include shape features derived from adaptive geometric descriptors based on eigenvectors and distribution features obtained through cylindrical coordinate transformation. Experimental results on real-world datasets validate the effectiveness of our method in various point cloud learning tasks, i.e., classification, part segmentation, and semantic segmentation.