Similarity search for videos based on robust latent semantic analysis

Kohei Inoue, Kiichi Urahama

Research output: Contribution to journalArticle

Abstract

A method retrieving videos is presented by utilizing vector quantization and latent semantic analysis. Each video is represented by a sequence of signatures through the vector quantization of frame datasets. Latent semantic analysis is then applied to the signature with a video matrix. We verified through experiments that dimensionality reduction in latent semantic analysis increases the speed and precision of retrieval. Making vector quantization more robust further improved the performance of similarity searches.

Original languageEnglish
Pages (from-to)1835-1838
Number of pages4
JournalKyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers
Volume58
Issue number12
DOIs
Publication statusPublished - Jan 1 2004

Fingerprint

Vector quantization
Semantics
Experiments

All Science Journal Classification (ASJC) codes

  • Media Technology
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Similarity search for videos based on robust latent semantic analysis. / Inoue, Kohei; Urahama, Kiichi.

In: Kyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers, Vol. 58, No. 12, 01.01.2004, p. 1835-1838.

Research output: Contribution to journalArticle

@article{62ee7305d77245a18d84ddcdee067c05,
title = "Similarity search for videos based on robust latent semantic analysis",
abstract = "A method retrieving videos is presented by utilizing vector quantization and latent semantic analysis. Each video is represented by a sequence of signatures through the vector quantization of frame datasets. Latent semantic analysis is then applied to the signature with a video matrix. We verified through experiments that dimensionality reduction in latent semantic analysis increases the speed and precision of retrieval. Making vector quantization more robust further improved the performance of similarity searches.",
author = "Kohei Inoue and Kiichi Urahama",
year = "2004",
month = "1",
day = "1",
doi = "10.3169/itej.58.1835",
language = "English",
volume = "58",
pages = "1835--1838",
journal = "Kyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers",
issn = "1342-6907",
publisher = "Institute of Image Information and Television Engineers",
number = "12",

}

TY - JOUR

T1 - Similarity search for videos based on robust latent semantic analysis

AU - Inoue, Kohei

AU - Urahama, Kiichi

PY - 2004/1/1

Y1 - 2004/1/1

N2 - A method retrieving videos is presented by utilizing vector quantization and latent semantic analysis. Each video is represented by a sequence of signatures through the vector quantization of frame datasets. Latent semantic analysis is then applied to the signature with a video matrix. We verified through experiments that dimensionality reduction in latent semantic analysis increases the speed and precision of retrieval. Making vector quantization more robust further improved the performance of similarity searches.

AB - A method retrieving videos is presented by utilizing vector quantization and latent semantic analysis. Each video is represented by a sequence of signatures through the vector quantization of frame datasets. Latent semantic analysis is then applied to the signature with a video matrix. We verified through experiments that dimensionality reduction in latent semantic analysis increases the speed and precision of retrieval. Making vector quantization more robust further improved the performance of similarity searches.

UR - http://www.scopus.com/inward/record.url?scp=12744256731&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=12744256731&partnerID=8YFLogxK

U2 - 10.3169/itej.58.1835

DO - 10.3169/itej.58.1835

M3 - Article

VL - 58

SP - 1835

EP - 1838

JO - Kyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers

JF - Kyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers

SN - 1342-6907

IS - 12

ER -