Anomalous Samples for Few-Shot Anomaly Detection

TL;DR

This paper explores how incorporating anomalous samples in few-shot anomaly detection improves model training, using multi-score techniques and augmentation-based validation to optimize detection performance on industrial datasets.

Contribution

It introduces a novel methodology that leverages anomalous samples with multi-score detection and augmentation validation, addressing the challenge of limited data in few-shot anomaly detection.

Findings

Anomalous samples can enhance detection performance in few-shot settings.

Multi-score anomaly detection improves robustness over single-score methods.

Augmentation-based validation optimizes score aggregation effectively.

Abstract

Several anomaly detection and classification methods rely on large amounts of non-anomalous or "normal" samples under the assump- tion that anomalous data is typically harder to acquire. This hypothesis becomes questionable in Few-Shot settings, where as little as one anno- tated sample can make a significant difference. In this paper, we tackle the question of utilizing anomalous samples in training a model for bi- nary anomaly classification. We propose a methodology that incorporates anomalous samples in a multi-score anomaly detection score leveraging recent Zero-Shot and memory-based techniques. We compare the utility of anomalous samples to that of regular samples and study the benefits and limitations of each. In addition, we propose an augmentation-based validation technique to optimize the aggregation of the different anomaly scores and demonstrate its effectiveness on popular industrial anomaly detection datasets.