Hierarchical Point-Patch Fusion with Adaptive Patch Codebook for 3D Shape Anomaly Detection

TL;DR

This paper introduces a hierarchical point-patch fusion network with an adaptive codebook for improved 3D shape anomaly detection, addressing generalization and noise sensitivity issues.

Contribution

It proposes a novel hierarchical network with adaptive patchification and self-supervised decomposition for robust, generalized 3D shape anomaly detection.

Findings

Achieves over 40% point-level improvement on industrial anomalies.

Improves average object-level detection by 7% on Real3D-AD.

Demonstrates superior AUC-ROC and AUC-PR performance on benchmarks.

Abstract

3D shape anomaly detection is a crucial task for industrial inspection and geometric analysis. Existing deep learning approaches typically learn representations of normal shapes and identify anomalies via out-of-distribution feature detection or decoder-based reconstruction. They often fail to generalize across diverse anomaly types and scales, such as global geometric errors (e.g., planar shifts, angle misalignments), and are sensitive to noisy or incomplete local points during training. To address these limitations, we propose a hierarchical point-patch anomaly scoring network that jointly models regional part features and local point features for robust anomaly reasoning. An adaptive patchification module integrates self-supervised decomposition to capture complex structural deviations. Beyond evaluations on public benchmarks (Anomaly-ShapeNet and Real3D-AD), we release an industrial test set with real CAD models exhibiting planar, angular, and structural defects. Experiments on public and industrial datasets show superior AUC-ROC and AUC-PR performance, including over 40% point-level improvement on the new industrial anomaly type and average object-level gains of 7% on Real3D-AD and 4% on Anomaly-ShapeNet, demonstrating strong robustness and generalization.