FastRef:Fast Prototype Refinement for Few-Shot Industrial Anomaly Detection

2025-06-26

AI Summary

In few-shot industrial anomaly detection (FS-IAD), weak representation capability of normal prototypes and susceptibility to anomaly contamination in query images hinder performance. To address this, we propose an iterative two-stage prototype refinement framework: Stage I employs a learnable feature transformation to transfer query-specific characteristics to the normal prototype; Stage II applies optimal transport (OT) to align prototype distributions, thereby suppressing anomaly-induced reconstruction bias. The method is compatible with linear reconstruction and can be seamlessly integrated—without architectural modification—into mainstream models including PatchCore, FastRecon, WinCLIP, and AnomalyDINO. Evaluated on four benchmarks (MVTec, ViSA, MPDD, RealIAD) under 1-/2-/4-shot settings, our approach significantly improves both pixel-level and image-level detection accuracy while maintaining efficient inference.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Feature Construction/Reformulation
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Few-shot industrial anomaly detection (FS-IAD) presents a critical challenge for practical automated inspection systems operating in data-scarce environments. While existing approaches predominantly focus on deriving prototypes from limited normal samples, they typically neglect to systematically incorporate query image statistics to enhance prototype representativeness. To address this issue, we propose FastRef, a novel and efficient prototype refinement framework for FS-IAD. Our method operates through an iterative two-stage process: (1) characteristic transfer from query features to prototypes via an optimizable transformation matrix, and (2) anomaly suppression through prototype alignment. The characteristic transfer is achieved through linear reconstruction of query features from prototypes, while the anomaly suppression addresses a key observation in FS-IAD that unlike conventional IAD with abundant normal prototypes, the limited-sample setting makes anomaly reconstruction more probable. Therefore, we employ optimal transport (OT) for non-Gaussian sampled features to measure and minimize the gap between prototypes and their refined counterparts for anomaly suppression. For comprehensive evaluation, we integrate FastRef with three competitive prototype-based FS-IAD methods: PatchCore, FastRecon, WinCLIP, and AnomalyDINO. Extensive experiments across four benchmark datasets of MVTec, ViSA, MPDD and RealIAD demonstrate both the effectiveness and computational efficiency of our approach under 1/2/4-shots.

Problem

• Enhancing prototype representativeness in few-shot anomaly detection
• Addressing anomaly reconstruction in limited-sample industrial settings
• Improving efficiency and accuracy of automated inspection systems

Innovation

• Iterative two-stage prototype refinement process
• Linear reconstruction for characteristic transfer
• Optimal transport for anomaly suppression