Joint Audio-Visual Words for Violent Scenes Detection in Movies - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

Joint Audio-Visual Words for Violent Scenes Detection in Movies

Résumé

This paper presents an audio-visual data representation for violent scenes detection in movies. Existing works in this field consider either the audio or the visual information; or their shallow fusion. None has yet explored their joint de- pendence for violent scenes detection. We propose a feature which provides strong multi-modal audio and visual cues by first joining the audio and the visual features and then revealing statistically the joint multi-modal patterns. Ex- perimental validation was conducted in the context of the Violent Scenes Detection task of the MediaEval 2013 Multi- media benchmark. The obtained results show the potential of the proposed approach in comparison to methods using audio and visual features separately and other fusion meth- ods.
Fichier non déposé

Dates et versions

hal-00981679 , version 1 (22-04-2014)

Identifiants

  • HAL Id : hal-00981679 , version 1

Citer

Nadia Derbas, Georges Quénot. Joint Audio-Visual Words for Violent Scenes Detection in Movies. International conference on multimedia retrieval, 2014, glasgow, United Kingdom. À paraître. ⟨hal-00981679⟩
51 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More