Audio-based Classification of Video Genre Using Multivariate Adaptive Regression Splines
Journal Title: International Journal of Advanced Research in Computer Engineering & Technology(IJARCET) - Year 2013, Vol 2, Issue 5
Abstract
A large number of researchers are attracted by video genre classification, video contents retrieval and semantics research in video processing and analysis domain. Many researchers try to propose structure or frameworks to classify the video genre that’s integrating many algorithms using low and high level features. Features generally include both useful and useless information that are difficult to separate. In this paper, video genre classification is proposed by using only the audio channel. A decomposition model is based on multivariate adaptive regression splines to separate useful and useless components and perform the genre identification is performed on these low-level acoustic features such as MFCC and timbral textual features. Experiments are conducted on a corpus composed from cartoons, sports, news, dahmas and musics on which obtain overall classification rate of 91.83%
Authors and Affiliations
Hnin Ei Latt , Dr. Nu War
Keywords
Related Articles
- EP ID EP115211
- DOI -
- Views 76
- Downloads 0
Fuzzy Inference System for Image Processing
Image Processing is one of the important applications of Computer Graphics. Presently applications require kinds of images are source of information for analysis. When image is converted from one form to another as prepr...
Data mining with Improved and efficient mechanism to detect the Vulnerabilities using intrusion detection system
Intrusion detection system must be capable of known and unknown vulnerabilities. We already studied the previous problems which includes detection of known vulnerabilities and unknown vulnerabilities. In order to obtai...
Quantify the intersection of behavior in the work environment
—This is the study of intersection of behaviour to analyze the behaviour of users in the social networking environment. We have so many opportunities and challenges to analyze the behaviour on a large scale. A so...
Conceptual Dependency Based Requirements Clustering for Component Selection
Component Based System (CBS) development is used to integrate existing components to build a software system, with the potential benefits of reducing development time and delivering quality system by using quality compon...
Optimized Surveillance Solution for Unattended Baggage Recognition
Abstract—The system automatically recognize activities around protected area to improve safety and security by multiplexing hundreds of video streams in real time. Object tracking method has important role in real tim...