Anomaly Detection and Segmentation (AD&S) is crucial for industrial quality control. While existing methods excel in generating anomaly scores for each pixel, practical applications require producing a binary segmentation to identify anomalies. Due to the absence of labeled anomalies in many real scenarios, standard practices binarize these maps based on some statistics derived from a validation set containing only nominal samples, resulting in poor segmentation performance. This paper addresses this problem by proposing a test time training strategy to improve the segmentation performance. Indeed, at test time, we can extract rich features directly from anomalous samples to train a classifier that can discriminate defects effectively. Our general approach can work downstream to any AD&S method that provides an anomaly score map as output, even in mul-timodal settings. We demonstrate the effectiveness of our approach over baselines through extensive experimentation and evaluation on MVTec AD and MVTec 3D-AD.

Costanzino A., Zama Ramirez P., Del Moro M., Aiezzo A., Lisanti G., Salti S., et al. (2024). Test Time Training for Industrial Anomaly Segmentation [10.1109/CVPRW63382.2024.00395].

Test Time Training for Industrial Anomaly Segmentation

Costanzino A.;Zama Ramirez P.;Lisanti G.;Salti S.;Di Stefano L.
2024

Abstract

Anomaly Detection and Segmentation (AD&S) is crucial for industrial quality control. While existing methods excel in generating anomaly scores for each pixel, practical applications require producing a binary segmentation to identify anomalies. Due to the absence of labeled anomalies in many real scenarios, standard practices binarize these maps based on some statistics derived from a validation set containing only nominal samples, resulting in poor segmentation performance. This paper addresses this problem by proposing a test time training strategy to improve the segmentation performance. Indeed, at test time, we can extract rich features directly from anomalous samples to train a classifier that can discriminate defects effectively. Our general approach can work downstream to any AD&S method that provides an anomaly score map as output, even in mul-timodal settings. We demonstrate the effectiveness of our approach over baselines through extensive experimentation and evaluation on MVTec AD and MVTec 3D-AD.
2024
2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
3910
3920
Costanzino A., Zama Ramirez P., Del Moro M., Aiezzo A., Lisanti G., Salti S., et al. (2024). Test Time Training for Industrial Anomaly Segmentation [10.1109/CVPRW63382.2024.00395].
Costanzino A.; Zama Ramirez P.; Del Moro M.; Aiezzo A.; Lisanti G.; Salti S.; Di Stefano L.
File in questo prodotto:
Eventuali allegati, non sono esposti

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/994976
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact