MixedTeacher : Knowledge Distillation for fast inference textural anomaly detection - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

MixedTeacher : Knowledge Distillation for fast inference textural anomaly detection

Simon Thomine
  • Fonction : Auteur
  • PersonId : 1262670
Hichem Snoussi
Mahmoud Soua
  • Fonction : Auteur
  • PersonId : 1262671

Résumé

For a very long time, unsupervised learning for anomaly detection has been at the heart of image processing research and a stepping stone for high performance industrial automation process. With the emergence of CNN, several methods have been proposed such as Autoencoders, GAN, deep feature extraction, etc. In this paper, we propose a new method based on the promising concept of knowledge distillation which consists of training a network (the student) on normal samples while considering the output of a larger pretrained network (the teacher). The main contributions of this paper are twofold: First, a reduced student architecture with optimal layer selection is proposed, then a new Student-Teacher architecture with network bias reduction combining two teachers is proposed in order to jointly enhance the performance of anomaly detection and its localization accuracy. The proposed texture anomaly detector has an outstanding capability to detect defects in any texture and a fast inference time compared to the SOTA methods.
Fichier principal
Vignette du fichier
MixedTeacherLatex.pdf (1.09 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04131472 , version 1 (16-06-2023)

Identifiants

  • HAL Id : hal-04131472 , version 1

Citer

Simon Thomine, Hichem Snoussi, Mahmoud Soua. MixedTeacher : Knowledge Distillation for fast inference textural anomaly detection. 2023 International Conference on Computer Vision Theory and Applications (VISAPP 2023), Feb 2023, Lisbonne, Portugal. ⟨hal-04131472⟩

Collections

UTT UTT-FULL-TEXT
6 Consultations
33 Téléchargements

Partager

Gmail Facebook X LinkedIn More