Data augmentation and pre-trained networks for extremely low data regimes unsupervised visual inspection

TL;DR

This paper compares three pre-trained feature-based methods for unsupervised visual anomaly detection in low-data regimes and demonstrates that data augmentation significantly improves their robustness in industrial inspection scenarios.

Contribution

It provides a comparative analysis of KNN, Mahalanobis, and PaDiM methods under limited data conditions and highlights the effectiveness of image space data augmentation.

Findings

All methods are robust to small sample sizes.

Data augmentation greatly enhances detection performance.

Pre-trained features are effective for low-data anomaly detection.

Abstract

The use of deep features coming from pre-trained neural networks for unsupervised anomaly detection purposes has recently gathered momentum in the computer vision field. In particular, industrial inspection applications can take advantage of such features, as demonstrated by the multiple successes of related methods on the MVTec Anomaly Detection (MVTec AD) dataset. These methods make use of neural networks pre-trained on auxiliary classification tasks such as ImageNet. However, to our knowledge, no comparative study of robustness to the low data regimes between these approaches has been conducted yet. For quality inspection applications, the handling of limited sample sizes may be crucial as large quantities of images are not available for small series. In this work, we aim to compare three approaches based on deep pre-trained features when varying the quantity of available data in MVTec AD: KNN, Mahalanobis, and PaDiM. We show that although these methods are mostly robust to small sample sizes, they still can benefit greatly from using data augmentation in the original image space, which allows to deal with very small production runs.