Explainable Anomaly Detection for Context Semantic Awareness - Laboratoire Interdisciplinaire des Sciences du Numérique Access content directly
Conference Poster Year : 2024

Explainable Anomaly Detection for Context Semantic Awareness

Abstract

We introduce an explainable neural network for Unsupervised Video Anomaly Detection (UVAD). Video Anomaly Detection (VAD) is a critical area of research with extensive applications, especially in surveillance and security systems deployed in public spaces such as roads, factories, and shopping malls. The significance of VAD lies in its capacity to enhance safety and security measures by identifying unusual events or activities without requiring manually labeled abnormal videos for training.
Fichier principal
Vignette du fichier
Demo__summary (1).pdf (2.61 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04521991 , version 1 (26-03-2024)

Identifiers

  • HAL Id : hal-04521991 , version 1

Cite

Hui Yang, Mostepha Redouane Khouadjia, Nacéra Bennacer Seghouani, Yue Ma, Serge Delmas. Explainable Anomaly Detection for Context Semantic Awareness. Workshop HyCHA (Hybridation Connaissances, Humain et Apprentissage Statistique), Mar 2024, Gif sur Yvette, France. ⟨hal-04521991⟩
41 View
9 Download

Share

Gmail Mastodon Facebook X LinkedIn More