Taming Anomalies with Down-Up Sampling Networks: Group Center Preserving Reconstruction for 3D Anomaly Detection

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Liang, Hanzhe, Zhang, Jie, Dai, Tao, Shen, Linlin, Wang, Jinbao, Gao, Can
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911157032321024
author Liang, Hanzhe
Zhang, Jie
Dai, Tao
Shen, Linlin
Wang, Jinbao
Gao, Can
author_facet Liang, Hanzhe
Zhang, Jie
Dai, Tao
Shen, Linlin
Wang, Jinbao
Gao, Can
contents Reconstruction-based methods have demonstrated very promising results for 3D anomaly detection. However, these methods face great challenges in handling high-precision point clouds due to the large scale and complex structure. In this study, a Down-Up Sampling Network (DUS-Net) is proposed to reconstruct high-precision point clouds for 3D anomaly detection by preserving the group center geometric structure. The DUS-Net first introduces a Noise Generation module to generate noisy patches, which facilitates the diversity of training data and strengthens the feature representation for reconstruction. Then, a Down-sampling Network (Down-Net) is developed to learn an anomaly-free center point cloud from patches with noise injection. Subsequently, an Up-sampling Network (Up-Net) is designed to reconstruct high-precision point clouds by fusing multi-scale up-sampling features. Our method leverages group centers for construction, enabling the preservation of geometric structure and providing a more precise point cloud. Extensive experiments demonstrate the effectiveness of our proposed method, achieving state-of-the-art (SOTA) performance with an Object-level AUROC of 79.9% and 79.5%, and a Point-level AUROC of 71.2% and 84.7% on the Real3D-AD and Anomaly-ShapeNet datasets, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2507_03903
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Taming Anomalies with Down-Up Sampling Networks: Group Center Preserving Reconstruction for 3D Anomaly Detection
Liang, Hanzhe
Zhang, Jie
Dai, Tao
Shen, Linlin
Wang, Jinbao
Gao, Can
Computer Vision and Pattern Recognition
68T10
I.4; I.5; J.6
Reconstruction-based methods have demonstrated very promising results for 3D anomaly detection. However, these methods face great challenges in handling high-precision point clouds due to the large scale and complex structure. In this study, a Down-Up Sampling Network (DUS-Net) is proposed to reconstruct high-precision point clouds for 3D anomaly detection by preserving the group center geometric structure. The DUS-Net first introduces a Noise Generation module to generate noisy patches, which facilitates the diversity of training data and strengthens the feature representation for reconstruction. Then, a Down-sampling Network (Down-Net) is developed to learn an anomaly-free center point cloud from patches with noise injection. Subsequently, an Up-sampling Network (Up-Net) is designed to reconstruct high-precision point clouds by fusing multi-scale up-sampling features. Our method leverages group centers for construction, enabling the preservation of geometric structure and providing a more precise point cloud. Extensive experiments demonstrate the effectiveness of our proposed method, achieving state-of-the-art (SOTA) performance with an Object-level AUROC of 79.9% and 79.5%, and a Point-level AUROC of 71.2% and 84.7% on the Real3D-AD and Anomaly-ShapeNet datasets, respectively.
title Taming Anomalies with Down-Up Sampling Networks: Group Center Preserving Reconstruction for 3D Anomaly Detection
topic Computer Vision and Pattern Recognition
68T10
I.4; I.5; J.6
url https://arxiv.org/abs/2507.03903