VisionADIndustrial

This dataset is based on the MVTec and VisA datasets. The evaluation score is calculated by calculating the mean of the pooled class scores across the MVTec and VIsA datasets. In other works, an algorithm is ran and tested on each class of the MVTec and VisA datasets, these values are summed, and then divided by 27 (total number of classes). This dataset allows a thorough benchmarking of industrial anomaly detection, by taking into account both the MVTec and VIsA classes. The dataset is used in the Transactions of Machine Learning Research paper: VisionAD, a software package of performant anomaly detection algorithms, and Proportion Localised, an interpretable metric. The dataset is used to undertake a thorough benchmarking of the available anomaly detection algorithms.