en
Proceedings chapter
Open access
English

Automatic Violence Scenes Detection: A Multi-Modal Approach

Presented at Santa Croce in Fossabanda, Pisa, Italy, September 1-2, 2011
Publication date2011
Abstract

In this working note, we propose a set of features and a classification scheme for detecting automatically violent scenes in movies. The features are extracted from audio, video, and subtitles modalities of the movies. In violent scenes classification, we found the following features relevant: the short time audio energy, motion component, and shot words rate.We classified the shots into violent and non-violent using naive Bayesian, Linear Discriminant Analysis (LDA), and Quadratic Discriminant Analysis (QDA) targeting to maximize the precision of the detection in the first two minutes of retrieved content.

Keywords
  • Violence
  • Audio feature extraction
  • Visual feature extraction
  • Text-based features
  • Subtitles
  • Violence scenes detection
  • Classification
Citation (ISO format)
GNINKOUN, Gabin, SOLEYMANI, Mohammad. Automatic Violence Scenes Detection: A Multi-Modal Approach. In: Working Notes Proceedings of the MediaEval 2011 Workshop. Santa Croce in Fossabanda, Pisa, Italy. [s.l.] : [s.n.], 2011.
Main files (1)
Proceedings chapter (Accepted version)
accessLevelPublic
Identifiers
  • PID : unige:74292
1079views
151downloads

Technical informations

Creation07/21/2015 5:12:00 PM
First validation07/21/2015 5:12:00 PM
Update time03/14/2023 11:29:26 PM
Status update03/14/2023 11:29:25 PM
Last indexation01/16/2024 6:36:10 PM
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack