Video semantic event/concept detection using a subspace-based multimedia data mining framework

Mei-Ling Shyu, Zongxing Xie, Min Chen, Shu Ching Chen

Research output: Contribution to journalArticle

104 Citations (Scopus)

Abstract

In this paper, a subspace-based multimedia data mining framework is proposed for video semantic analysis, specifically video event/concept detection, by addressing two basic issues, i.e., semantic gap and rare event/concept detection. The proposed framework achieves full automation via multimodal content analysis and intelligent integration of distance-based and rule-based data mining techniques. The content analysis process facilitates the comprehensive video analysis by extracting low-level and middle-level features from audio/visual channels. The integrated data mining techniques effectively address these two basic issues by alleviating the class imbalance issue along the process and by reconstructing and refining the feature dimension automatically. The promising experimental performance on goal/corner event detection and sports/commercials/building concepts extraction from soccer videos and TRECVID news collections demonstrates the effectiveness of the proposed framework. Furthermore, its unique domain-free characteristic indicates the great potential of extending the proposed multimedia data mining framework to a wide range of different application domains.

Original languageEnglish
Pages (from-to)252-259
Number of pages8
JournalIEEE Transactions on Multimedia
Volume10
Issue number2
DOIs
StatePublished - Feb 1 2008

Fingerprint

Data mining
Semantics
Sports
Refining
Automation

Keywords

  • Data mining
  • Eigenspace
  • Eigenvalue
  • Event/concept detection
  • Principal component
  • Video semantics analysis

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

Video semantic event/concept detection using a subspace-based multimedia data mining framework. / Shyu, Mei-Ling; Xie, Zongxing; Chen, Min; Chen, Shu Ching.

In: IEEE Transactions on Multimedia, Vol. 10, No. 2, 01.02.2008, p. 252-259.

Research output: Contribution to journalArticle

Shyu, Mei-Ling ; Xie, Zongxing ; Chen, Min ; Chen, Shu Ching. / Video semantic event/concept detection using a subspace-based multimedia data mining framework. In: IEEE Transactions on Multimedia. 2008 ; Vol. 10, No. 2. pp. 252-259.
@article{fb55a87666c84ab5ae7719ed5ea33ac2,
title = "Video semantic event/concept detection using a subspace-based multimedia data mining framework",
abstract = "In this paper, a subspace-based multimedia data mining framework is proposed for video semantic analysis, specifically video event/concept detection, by addressing two basic issues, i.e., semantic gap and rare event/concept detection. The proposed framework achieves full automation via multimodal content analysis and intelligent integration of distance-based and rule-based data mining techniques. The content analysis process facilitates the comprehensive video analysis by extracting low-level and middle-level features from audio/visual channels. The integrated data mining techniques effectively address these two basic issues by alleviating the class imbalance issue along the process and by reconstructing and refining the feature dimension automatically. The promising experimental performance on goal/corner event detection and sports/commercials/building concepts extraction from soccer videos and TRECVID news collections demonstrates the effectiveness of the proposed framework. Furthermore, its unique domain-free characteristic indicates the great potential of extending the proposed multimedia data mining framework to a wide range of different application domains.",
keywords = "Data mining, Eigenspace, Eigenvalue, Event/concept detection, Principal component, Video semantics analysis",
author = "Mei-Ling Shyu and Zongxing Xie and Min Chen and Chen, {Shu Ching}",
year = "2008",
month = "2",
day = "1",
doi = "10.1109/TMM.2007.911830",
language = "English",
volume = "10",
pages = "252--259",
journal = "IEEE Transactions on Multimedia",
issn = "1520-9210",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

TY - JOUR

T1 - Video semantic event/concept detection using a subspace-based multimedia data mining framework

AU - Shyu, Mei-Ling

AU - Xie, Zongxing

AU - Chen, Min

AU - Chen, Shu Ching

PY - 2008/2/1

Y1 - 2008/2/1

N2 - In this paper, a subspace-based multimedia data mining framework is proposed for video semantic analysis, specifically video event/concept detection, by addressing two basic issues, i.e., semantic gap and rare event/concept detection. The proposed framework achieves full automation via multimodal content analysis and intelligent integration of distance-based and rule-based data mining techniques. The content analysis process facilitates the comprehensive video analysis by extracting low-level and middle-level features from audio/visual channels. The integrated data mining techniques effectively address these two basic issues by alleviating the class imbalance issue along the process and by reconstructing and refining the feature dimension automatically. The promising experimental performance on goal/corner event detection and sports/commercials/building concepts extraction from soccer videos and TRECVID news collections demonstrates the effectiveness of the proposed framework. Furthermore, its unique domain-free characteristic indicates the great potential of extending the proposed multimedia data mining framework to a wide range of different application domains.

AB - In this paper, a subspace-based multimedia data mining framework is proposed for video semantic analysis, specifically video event/concept detection, by addressing two basic issues, i.e., semantic gap and rare event/concept detection. The proposed framework achieves full automation via multimodal content analysis and intelligent integration of distance-based and rule-based data mining techniques. The content analysis process facilitates the comprehensive video analysis by extracting low-level and middle-level features from audio/visual channels. The integrated data mining techniques effectively address these two basic issues by alleviating the class imbalance issue along the process and by reconstructing and refining the feature dimension automatically. The promising experimental performance on goal/corner event detection and sports/commercials/building concepts extraction from soccer videos and TRECVID news collections demonstrates the effectiveness of the proposed framework. Furthermore, its unique domain-free characteristic indicates the great potential of extending the proposed multimedia data mining framework to a wide range of different application domains.

KW - Data mining

KW - Eigenspace

KW - Eigenvalue

KW - Event/concept detection

KW - Principal component

KW - Video semantics analysis

UR - http://www.scopus.com/inward/record.url?scp=38349183005&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38349183005&partnerID=8YFLogxK

U2 - 10.1109/TMM.2007.911830

DO - 10.1109/TMM.2007.911830

M3 - Article

AN - SCOPUS:38349183005

VL - 10

SP - 252

EP - 259

JO - IEEE Transactions on Multimedia

JF - IEEE Transactions on Multimedia

SN - 1520-9210

IS - 2

ER -