UNSUPERVISED AND ADAPTIVE PERIMETER INTRUSION DETECTOR - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

UNSUPERVISED AND ADAPTIVE PERIMETER INTRUSION DETECTOR

Quentin Barthélemy
  • Fonction : Auteur
  • PersonId : 1080669
Sarah Bertrand
Laure Tougne

Résumé

Perimeter intrusion detection (PID) deals with the detection of intruders displacing in a protected perimeter. In the video surveillance domain, deep learning has shown tremendous progresses. Existing deep learning based PID systems (PIDS) are supervised and thus require a lot of annotated data. However, since intrusions are rare events, there are very few positives in datasets, thus making them highly imbalanced. Furthermore, a PIDS must adapt to varying real-life scene dynamics, like weather, light, environmental conditions, etc. To address these issues, we propose an autoencoder-based, endto-end trainable, unsupervised PIDS with a module that can adapt to long-term variations in scene dynamics. Our results show competitive performance of the proposed system on the standard i-LIDS dataset.
Fichier principal
Vignette du fichier
lohani_pids_ICIP_22.pdf (954.56 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03800865 , version 1 (06-10-2022)

Identifiants

Citer

Devashish Lohani, Carlos F Crispim-Junior, Quentin Barthélemy, Sarah Bertrand, Lionel Robinault, et al.. UNSUPERVISED AND ADAPTIVE PERIMETER INTRUSION DETECTOR. IEEE International Conference on Image Processing (ICIP), Oct 2022, Bordeaux, France. ⟨10.1109/ICIP46576.2022.9897472⟩. ⟨hal-03800865⟩
35 Consultations
97 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More