Indexing technique for similarity matching in large video databases

Sanghyun Park, June Suh Cho, Ki Ho Hyun

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Similarity matching in video databases is of growing importance in many new applications such as video clustering and digital video libraries. In order to provide efficient access to relevant data in large databases, there have been many research efforts in video indexing with diverse spatial and temporal features. However, most of the previous works relied on sequential matching methods or memory-based inverted file techniques, thus making them unsuitable for a large volume of video databases. In order to resolve this problem, this paper proposes an effective and scalable indexing technique using a trie, originally proposed for string matching, as an index structure. For building an index, we convert each frame into a symbol sequence using a window order heuristic and build a disk-resident trie from a set of symbol sequences. For query processing, we perform a depth-first search on the trie and execute a temporal segmentation. To verify the superiority of our approach, we perform several experiments with real and synthetic data sets. The results reveal that our approach consistently outperforms the sequential scan method, and the performance gain is maintained even with a large volume of video databases.

Original languageEnglish
Pages (from-to)214-222
Number of pages9
JournalProceedings of SPIE - The International Society for Optical Engineering
Volume4676
DOIs
Publication statusPublished - 2002 Jan 1

Fingerprint

Video Databases
Indexing
Video Indexing
Depth-first Search
String Matching
Digital Video
Query Processing
Synthetic Data
files
Convert
Query processing
Resolve
strings
Segmentation
Clustering
Heuristics
Verify
Data storage equipment
Experiment
Similarity

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

@article{de3b4cfd68514673984b8a7506d1c2d8,
title = "Indexing technique for similarity matching in large video databases",
abstract = "Similarity matching in video databases is of growing importance in many new applications such as video clustering and digital video libraries. In order to provide efficient access to relevant data in large databases, there have been many research efforts in video indexing with diverse spatial and temporal features. However, most of the previous works relied on sequential matching methods or memory-based inverted file techniques, thus making them unsuitable for a large volume of video databases. In order to resolve this problem, this paper proposes an effective and scalable indexing technique using a trie, originally proposed for string matching, as an index structure. For building an index, we convert each frame into a symbol sequence using a window order heuristic and build a disk-resident trie from a set of symbol sequences. For query processing, we perform a depth-first search on the trie and execute a temporal segmentation. To verify the superiority of our approach, we perform several experiments with real and synthetic data sets. The results reveal that our approach consistently outperforms the sequential scan method, and the performance gain is maintained even with a large volume of video databases.",
author = "Sanghyun Park and Cho, {June Suh} and Hyun, {Ki Ho}",
year = "2002",
month = "1",
day = "1",
doi = "10.1117/12.451093",
language = "English",
volume = "4676",
pages = "214--222",
journal = "Proceedings of SPIE - The International Society for Optical Engineering",
issn = "0277-786X",
publisher = "SPIE",

}

Indexing technique for similarity matching in large video databases. / Park, Sanghyun; Cho, June Suh; Hyun, Ki Ho.

In: Proceedings of SPIE - The International Society for Optical Engineering, Vol. 4676, 01.01.2002, p. 214-222.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Indexing technique for similarity matching in large video databases

AU - Park, Sanghyun

AU - Cho, June Suh

AU - Hyun, Ki Ho

PY - 2002/1/1

Y1 - 2002/1/1

N2 - Similarity matching in video databases is of growing importance in many new applications such as video clustering and digital video libraries. In order to provide efficient access to relevant data in large databases, there have been many research efforts in video indexing with diverse spatial and temporal features. However, most of the previous works relied on sequential matching methods or memory-based inverted file techniques, thus making them unsuitable for a large volume of video databases. In order to resolve this problem, this paper proposes an effective and scalable indexing technique using a trie, originally proposed for string matching, as an index structure. For building an index, we convert each frame into a symbol sequence using a window order heuristic and build a disk-resident trie from a set of symbol sequences. For query processing, we perform a depth-first search on the trie and execute a temporal segmentation. To verify the superiority of our approach, we perform several experiments with real and synthetic data sets. The results reveal that our approach consistently outperforms the sequential scan method, and the performance gain is maintained even with a large volume of video databases.

AB - Similarity matching in video databases is of growing importance in many new applications such as video clustering and digital video libraries. In order to provide efficient access to relevant data in large databases, there have been many research efforts in video indexing with diverse spatial and temporal features. However, most of the previous works relied on sequential matching methods or memory-based inverted file techniques, thus making them unsuitable for a large volume of video databases. In order to resolve this problem, this paper proposes an effective and scalable indexing technique using a trie, originally proposed for string matching, as an index structure. For building an index, we convert each frame into a symbol sequence using a window order heuristic and build a disk-resident trie from a set of symbol sequences. For query processing, we perform a depth-first search on the trie and execute a temporal segmentation. To verify the superiority of our approach, we perform several experiments with real and synthetic data sets. The results reveal that our approach consistently outperforms the sequential scan method, and the performance gain is maintained even with a large volume of video databases.

UR - http://www.scopus.com/inward/record.url?scp=0036029058&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036029058&partnerID=8YFLogxK

U2 - 10.1117/12.451093

DO - 10.1117/12.451093

M3 - Article

AN - SCOPUS:0036029058

VL - 4676

SP - 214

EP - 222

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

SN - 0277-786X

ER -