Unveiling the Anomalies in an Ever-Changing World: A Benchmark for Pixel-Level Anomaly Detection in Continual Learning

arXiv:2403.15463v1 Announce Type: cross
Abstract: Anomaly Detection is a relevant problem in numerous real-world applications, especially when dealing with images. However, little attention has been paid to the issue of changes over time in the input data distribution, which may cause a significant decrease in performance. In this study, we investigate the problem of Pixel-Level Anomaly Detection in the Continual Learning setting, where new data arrives over time and the goal is to perform well on new and old data. …

abstract anomaly anomaly detection applications arxiv attention benchmark continual cs.cv cs.lg data detection distribution however images issue pixel type world