Please use this identifier to cite or link to this item: http://dx.doi.org/10.14279/depositonce-6799
Main Title: Detecting violent content in Hollywood movies by mid-level audio representations
Author(s): Acar, Esra
Hopfgartner, Frank
Alyabrak, Sahin
Type: Conference Object
Language Code: en
Abstract: Movie violent content detection e.g., for providing automated youth protection services is a valuable video content analysis functionality. Choosing discriminative features for the representation of video segments is a key issue in designing violence detection algorithms. In this paper, we employ mid-level audio features which are based on a Bag-of-Audio Words (BoAW) method using Mel-Frequency Cepstral Coefficients (MFCC). BoAW representations are constructed with two different meth- ods, namely the vector quantization-based (VQ-based) method and the sparse coding-based (SC-based) method. We choose two- class support vector machines (SVMs) for classifying video shots as (non-)violent. Our experimental results on detecting violent video shots in Hollywood movies show that the mid-level audio features provide promising results. Additionally, we establish that the SC-based method outperforms the VQ-based one. More importantly, the SC-based method outperforms the unimodal submissions in the MediaEval Violent Scenes Detection (VSD) task except one visual-based method in terms of average precision.
URI: https://depositonce.tu-berlin.de//handle/11303/7609
http://dx.doi.org/10.14279/depositonce-6799
Issue Date: 2013
Date Available: 17-Apr-2018
DDC Class: 000 Informatik, Informationswissenschaft, allgemeine Werke
Subject(s): video coding
feature extraction
mel frequency cepstral coefficient
visualization
support vector machines
dictionaries
training
License: http://rightsstatements.org/vocab/InC/1.0/
Proceedings Title: 2013 11th International Workshop on Content-Based Multimedia Indexing (CBMI)
Publisher: IEEE
Publisher Place: Veszprem, Hungary
Volume: 2013
Publisher DOI: 10.1109/CBMI.2013.6576556
Page Start: 73
Page End: 78
Series: Proceedings of the 11th International Workshop on Content-Based Multimedia Indexing (CBMI)
ISBN: 978-1-4799-0956-8
ISSN: 1949-3991
Appears in Collections:FG Agententechnologien in betrieblichen Anwendungen und der Telekommunikation (AOT) » Publications

Files in This Item:
File Description SizeFormat 
2013_acar_etal.pdf2.99 MBAdobe PDFThumbnail
View/Open


Items in DepositOnce are protected by copyright, with all rights reserved, unless otherwise indicated.