3D-SONAR: Self-Organizing Network for 3D Anomaly Ranking

2026-01-14

AI Summary

This work addresses the challenge of 3D point cloud surface anomaly detection in industrial inspection, which typically relies on large-scale annotated data. We propose the first unsupervised method that integrates self-organizing networks with a physics-inspired mechanism. By modeling the point cloud as a dynamic system and constructing an undirected graph embedded with attractive and repulsive interactions, our approach leverages the resulting energy distribution to rank anomalies without any training. The method requires no labeled or unlabeled training data and demonstrates strong detection performance on both open and closed surfaces, confirming its effectiveness and generalization capability across diverse geometric configurations.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Computer Vision: 3D Computer Vision

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Surface anomaly detection using 3D point cloud data has gained increasing attention in industrial inspection. However, most existing methods rely on deep learning techniques that are highly dependent on large-scale datasets for training, which are difficult and expensive to acquire in real-world applications. To address this challenge, we propose a novel method based on self-organizing network for 3D anomaly ranking, also named 3D-SONAR. The core idea is to model the 3D point cloud as a dynamic system, where the points are represented as an undirected graph and interact via attractive and repulsive forces. The energy distribution induced by these forces can reveal surface anomalies. Experimental results show that our method achieves superior anomaly detection performance in both open surface and closed surface without training. This work provides a new perspective on unsupervised inspection and highlights the potential of physics-inspired models in industrial anomaly detection tasks with limited data.

Problem

• 3D point cloud
• anomaly detection
• industrial inspection
• limited data
• unsupervised learning

Innovation

• self-organizing network
• 3D anomaly detection
• physics-inspired model
• unsupervised learning
• point cloud