MC3D-AD: A Unified Geometry-aware Reconstruction Model for Multi-category 3D Anomaly Detection

2025-05-04

AI Summary

Existing 3D anomaly detection methods require class-specific model training, resulting in poor generalization and high deployment costs. This paper proposes the first unified unsupervised 3D anomaly detection framework for multi-class industrial quality inspection. It jointly models local–global geometric structures of normal samples and enables cross-class reconstruction via a geometry-aware masked attention mechanism, a local-grouping encoder, and a position-embedding-enhanced global query decoder. Key innovations include: (1) adaptive geometry-aware masked attention; (2) a Transformer-based decoder explicitly incorporating point cloud positional information; and (3) an unsupervised anomaly scoring scheme based on reconstruction error. Evaluated on Real3D-AD and Anomaly-ShapeNet, our method achieves object-level AUROC improvements of 3.1% and 9.3%, respectively—significantly outperforming state-of-the-art single-class approaches.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

3D Anomaly Detection (AD) is a promising means of controlling the quality of manufactured products. However, existing methods typically require carefully training a task-specific model for each category independently, leading to high cost, low efficiency, and weak generalization. Therefore, this paper presents a novel unified model for Multi-Category 3D Anomaly Detection (MC3D-AD) that aims to utilize both local and global geometry-aware information to reconstruct normal representations of all categories. First, to learn robust and generalized features of different categories, we propose an adaptive geometry-aware masked attention module that extracts geometry variation information to guide mask attention. Then, we introduce a local geometry-aware encoder reinforced by the improved mask attention to encode group-level feature tokens. Finally, we design a global query decoder that utilizes point cloud position embeddings to improve the decoding process and reconstruction ability. This leads to local and global geometry-aware reconstructed feature tokens for the AD task. MC3D-AD is evaluated on two publicly available Real3D-AD and Anomaly-ShapeNet datasets, and exhibits significant superiority over current state-of-the-art single-category methods, achieving 3.1% and 9.3% improvement in object-level AUROC over Real3D-AD and Anomaly-ShapeNet, respectively. The source code will be released upon acceptance.

Problem

• Existing methods require task-specific models per category, increasing costs and reducing efficiency.
• Current approaches lack generalization in multi-category 3D anomaly detection tasks.
• Local and global geometry-aware reconstruction is needed for accurate anomaly detection.

Innovation

• Adaptive geometry-aware masked attention module
• Local geometry-aware encoder with improved attention
• Global query decoder with position embeddings