A Metacognitive Perspective on Anomaly Detection with Vision Foundation Models
Proceedings of the AAAI Symposium Series · 06.11.2026 · 10 (1) · 110-114
DOI: 10.1609/aaaiss.v10i1.43397 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Symposium Series, online erschienen am 06.11.2026. Verfasst von Matteo Galiazzo und 4 weiteren Autor:innen.
Abstract (Originalsprache)
Anomaly Detection (AD) in images identifies defective regions by scoring against a model of normality learned from defect-free data. Current state of the art methods encode images via a frozen Vision Foundation Model (VFM) backbone and score them by their Nearest-Neighbor (NN) distance to a memory bank of normal samples. Recently, due to the parallel evolution of VFM encoders and scoring mechanism, the need to disentangle their individual contributions has emerged, in order to guarantee the deployment of the best performing model in specific industrial production environments. In our work, we analyze AD within a metacognitive perspective, we freeze the backbone and swap the scorer, replacing NN distance with a Mahalanobis score over a Gaussian distribution. By doing that, we can evaluate two different metacognitive functions of the scoring mechanisms: monitoring how far an observation is from the learned model of normality via a threshold-free metric, and deciding when the difference justifies an action according to a threshold. We show that these two metacognitive functions can dissociate. We further notice that replacing a NN memory bank with a simpler Mahalanobis model improves pixel-level monitoring (+8.00 AUPRO_0.05 on MVTec AD 2) while degrading the segmentation performance after the thresholding step (-13.46 F_1). Lastly, we show that a threshold derived by Extreme Value Theory (EVT) recovers that loss without using anomalous data: the decision stops contradicting the monitoring signal, the F_1 difference between the two scorers decreases from 13.46 to 0.61 points on MVTec AD 2, with consistent gains on MVTec AD and VisA.
Zitieren
APA
Galiazzo, M., Onorato, G., Loi, D., Bastian, N. D., & Restuccia, F. (2026). A Metacognitive Perspective on Anomaly Detection with Vision Foundation Models. Proceedings of the AAAI Symposium Series, 10(1), 110-114. https://doi.org/10.1609/aaaiss.v10i1.43397
Vancouver
Galiazzo M, Onorato G, Loi D, Bastian ND, Restuccia F. A Metacognitive Perspective on Anomaly Detection with Vision Foundation Models. Proceedings of the AAAI Symposium Series. 2026;10(1):110-114. doi:10.1609/aaaiss.v10i1.43397
BibTeX
@article{Galiazzo2026,
author = {Galiazzo, Matteo and Onorato, Gabriele and Loi, Dario and Bastian, Nathaniel D. and Restuccia, Francesco},
title = {A Metacognitive Perspective on Anomaly Detection with Vision Foundation Models},
journal = {Proceedings of the AAAI Symposium Series},
year = {2026},
volume = {10},
number = {1},
pages = {110--114},
doi = {10.1609/aaaiss.v10i1.43397}
}
EUVIMED