A Point Cloud Anomaly Detection Method and A DataSet

1 Huazhong University of Science and Technology, 2 Hunan University
UnderReview

*Equal Contribution    # Co-corresponding author

Abstract

In industrial point cloud analysis, detecting subtle anomalies demands high-resolution spatial data, yet prevailing benchmarks emphasize low-resolution inputs. To address this disparity, we propose a scalable pipeline for generating realistic and subtle 3D anomalies. Employing this pipeline, we developed MiniShift, the inaugural high-resolution 3D anomaly detection dataset, encompassing 2,577 point clouds, each with 500,000 points and anomalies occupying less than 1\% of the total. We further introduce Simple3D, an efficient framework integrating Multi-scale Neighborhood Descriptors (MSND) and Local Feature Spatial Aggregation (LFSA) to capture intricate geometric details with minimal computational overhead, achieving real-time inference exceeding 20 fps. Extensive evaluations on MiniShift and established benchmarks demonstrate that Simple3D surpasses state-of-the-art methods in both accuracy and speed, highlighting the pivotal role of high-resolution data and effective feature aggregation in advancing practical 3D anomaly detection.

MiniShift Dataset
111
Simple3D
111
Results on MiniShift
111
Visualization on MiniShift
111
Results on Three Public Datasets
111
Visualization on Three Public Datasets
111

BibTeX

BibTex Code Here