Correlation-based video semantic concept detection using multiple correspondence analysis

Lin Lin, Guy Ravitz, Mei-Ling Shyu, Shu Ching Chen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

37 Citations (Scopus)

Abstract

Semantic concept detection has emerged as an intriguing topic in multimedia research recently. The ability to interpret high-level semantics from low-level features has been the long desired goal of many researchers. In this paper, we propose a novel framework that utilizes the ability of multiple correspondence analysis (MCA) to explore the correlation between different items (feature-value pairs) and classes (concepts) to bridge the gap between the extracted low-level features and high-level semantic concepts. Using the concepts and benchmark data identified and provided by the TRECVID project, we have shown that our proposed framework demonstrates promising results and performs better than the Decision Tree (DT), Support Vector Machine (SVM), and Naive Bayesian (NB) classifiers that are commonly applied to the TRECVID datasets.

Original languageEnglish
Title of host publicationProceedings - 10th IEEE International Symposium on Multimedia, ISM 2008
Pages316-321
Number of pages6
DOIs
StatePublished - Dec 1 2008
Event10th IEEE International Symposium on Multimedia, ISM 2008 - Berkeley, CA, United States
Duration: Dec 15 2008Dec 17 2008

Other

Other10th IEEE International Symposium on Multimedia, ISM 2008
CountryUnited States
CityBerkeley, CA
Period12/15/0812/17/08

Fingerprint

Semantics
Decision trees
Support vector machines
Classifiers

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Lin, L., Ravitz, G., Shyu, M-L., & Chen, S. C. (2008). Correlation-based video semantic concept detection using multiple correspondence analysis. In Proceedings - 10th IEEE International Symposium on Multimedia, ISM 2008 (pp. 316-321). [4741186] https://doi.org/10.1109/ISM.2008.111

Correlation-based video semantic concept detection using multiple correspondence analysis. / Lin, Lin; Ravitz, Guy; Shyu, Mei-Ling; Chen, Shu Ching.

Proceedings - 10th IEEE International Symposium on Multimedia, ISM 2008. 2008. p. 316-321 4741186.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lin, L, Ravitz, G, Shyu, M-L & Chen, SC 2008, Correlation-based video semantic concept detection using multiple correspondence analysis. in Proceedings - 10th IEEE International Symposium on Multimedia, ISM 2008., 4741186, pp. 316-321, 10th IEEE International Symposium on Multimedia, ISM 2008, Berkeley, CA, United States, 12/15/08. https://doi.org/10.1109/ISM.2008.111
Lin L, Ravitz G, Shyu M-L, Chen SC. Correlation-based video semantic concept detection using multiple correspondence analysis. In Proceedings - 10th IEEE International Symposium on Multimedia, ISM 2008. 2008. p. 316-321. 4741186 https://doi.org/10.1109/ISM.2008.111
Lin, Lin ; Ravitz, Guy ; Shyu, Mei-Ling ; Chen, Shu Ching. / Correlation-based video semantic concept detection using multiple correspondence analysis. Proceedings - 10th IEEE International Symposium on Multimedia, ISM 2008. 2008. pp. 316-321
@inproceedings{9d2eab80a2be4aa4ba292423a4f1345f,
title = "Correlation-based video semantic concept detection using multiple correspondence analysis",
abstract = "Semantic concept detection has emerged as an intriguing topic in multimedia research recently. The ability to interpret high-level semantics from low-level features has been the long desired goal of many researchers. In this paper, we propose a novel framework that utilizes the ability of multiple correspondence analysis (MCA) to explore the correlation between different items (feature-value pairs) and classes (concepts) to bridge the gap between the extracted low-level features and high-level semantic concepts. Using the concepts and benchmark data identified and provided by the TRECVID project, we have shown that our proposed framework demonstrates promising results and performs better than the Decision Tree (DT), Support Vector Machine (SVM), and Naive Bayesian (NB) classifiers that are commonly applied to the TRECVID datasets.",
author = "Lin Lin and Guy Ravitz and Mei-Ling Shyu and Chen, {Shu Ching}",
year = "2008",
month = "12",
day = "1",
doi = "10.1109/ISM.2008.111",
language = "English",
isbn = "9780769534541",
pages = "316--321",
booktitle = "Proceedings - 10th IEEE International Symposium on Multimedia, ISM 2008",

}

TY - GEN

T1 - Correlation-based video semantic concept detection using multiple correspondence analysis

AU - Lin, Lin

AU - Ravitz, Guy

AU - Shyu, Mei-Ling

AU - Chen, Shu Ching

PY - 2008/12/1

Y1 - 2008/12/1

N2 - Semantic concept detection has emerged as an intriguing topic in multimedia research recently. The ability to interpret high-level semantics from low-level features has been the long desired goal of many researchers. In this paper, we propose a novel framework that utilizes the ability of multiple correspondence analysis (MCA) to explore the correlation between different items (feature-value pairs) and classes (concepts) to bridge the gap between the extracted low-level features and high-level semantic concepts. Using the concepts and benchmark data identified and provided by the TRECVID project, we have shown that our proposed framework demonstrates promising results and performs better than the Decision Tree (DT), Support Vector Machine (SVM), and Naive Bayesian (NB) classifiers that are commonly applied to the TRECVID datasets.

AB - Semantic concept detection has emerged as an intriguing topic in multimedia research recently. The ability to interpret high-level semantics from low-level features has been the long desired goal of many researchers. In this paper, we propose a novel framework that utilizes the ability of multiple correspondence analysis (MCA) to explore the correlation between different items (feature-value pairs) and classes (concepts) to bridge the gap between the extracted low-level features and high-level semantic concepts. Using the concepts and benchmark data identified and provided by the TRECVID project, we have shown that our proposed framework demonstrates promising results and performs better than the Decision Tree (DT), Support Vector Machine (SVM), and Naive Bayesian (NB) classifiers that are commonly applied to the TRECVID datasets.

UR - http://www.scopus.com/inward/record.url?scp=62949200344&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=62949200344&partnerID=8YFLogxK

U2 - 10.1109/ISM.2008.111

DO - 10.1109/ISM.2008.111

M3 - Conference contribution

SN - 9780769534541

SP - 316

EP - 321

BT - Proceedings - 10th IEEE International Symposium on Multimedia, ISM 2008

ER -