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

TL;DR

FastRef introduces an efficient prototype refinement framework for few-shot industrial anomaly detection, leveraging query image statistics and optimal transport to improve prototype quality and anomaly detection accuracy in data-scarce settings.

Contribution

The paper proposes a novel iterative prototype refinement method using characteristic transfer and anomaly suppression with optimal transport, enhancing few-shot anomaly detection performance.

Findings

Effective across multiple benchmark datasets.

Improves detection accuracy in 1/2/4-shot scenarios.

Demonstrates computational efficiency and robustness.

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.