Search Results for author: Thomas Defard

Found 1 papers, 1 papers with code

PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization

22 code implementations17 Nov 2020 Thomas Defard, Aleksandr Setkov, Angelique Loesch, Romaric Audigier

We present a new framework for Patch Distribution Modeling, PaDiM, to concurrently detect and localize anomalies in images in a one-class learning setting.

 Ranked #1 on on MVTecAD

Unsupervised Anomaly Detection

Cannot find the paper you are looking for? You can Submit a new open access paper.