MAESTRO: A Full Point Cloud Approach for 3D Anomaly Detection Based on Reconstruction

Remi Lhoste, Remi Lhoste, Antoine Vacavant, Damien Delhay

2025

Abstract

3D anomaly detection is a critical task in industrial manufacturing, for maintaining product quality and operational safety. However, many existing methods function more as 2.5D anomaly detection techniques, primarily relying on image data and underexploiting point clouds. These methods often face challenges related to real scenarios, and reliance on large pretrained models or memory banks. To address these issues, we propose MAESTRO, a Masked AutoEncoder Self-Supervised Through Reconstruction Only. This novel 3D anomaly detection method based solely on point cloud reconstruction without utilizing pretrained models or memory banks, making it particularly suitable for industrial applications. Experiments demonstrate that our method can outperform previous state-of-the-art methods on several classes of the MVTec 3D-AD dataset (Bergmann et al., 2022).

Paper Citation

in Harvard Style

Lhoste R., Vacavant A. and Delhay D. (2025). MAESTRO: A Full Point Cloud Approach for 3D Anomaly Detection Based on Reconstruction. In Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP; ISBN 978-989-758-728-3, SciTePress, pages 717-724. DOI: 10.5220/0013250500003912

in Bibtex Style

@conference{visapp25,
author={Remi Lhoste and Antoine Vacavant and Damien Delhay},
title={MAESTRO: A Full Point Cloud Approach for 3D Anomaly Detection Based on Reconstruction},
booktitle={Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP},
year={2025},
pages={717-724},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013250500003912},
isbn={978-989-758-728-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP
TI - MAESTRO: A Full Point Cloud Approach for 3D Anomaly Detection Based on Reconstruction
SN - 978-989-758-728-3
AU - Lhoste R.
AU - Vacavant A.
AU - Delhay D.
PY - 2025
SP - 717
EP - 724
DO - 10.5220/0013250500003912
PB - SciTePress