Model-based semantic object extraction for content-based video representation and indexing

J. Fan, D. K Y Yau, M. S. Hacid, Ahmed Elmagarmid

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

3 Citations (Scopus)

Abstract

This paper proposes an integrated system for supporting content-based video retrieval and browsing over networks. An automatic semantic video object extraction technique for providing more compact video representation is developed. The video images are first partitioned into a set of homogeneous regions with accurate boundaries by integrating the results of color edge detection and region growing procedures. The object seeds, which are the intuitive and representative part of the semantic objects, are detected from these obtained homogeneous image regions. The semantic objects are then generated by a seeded region aggregation or a human interaction procedure. These obtained semantic objects are tracked along the time axis for exploiting their temporal correspondences among frames. Given the semantic video objects represented by a set of visual features, a seeded semantic video content clustering technique is developed for providing more effective video indexing, retrieval and browsing.

Original languageEnglish
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsM.M Yeung, C. Li, R.W. Lienhart
Pages523-535
Number of pages13
Volume4315
DOIs
Publication statusPublished - 2001
Externally publishedYes
EventStorage and Retrieval for Media Databases 2001 - San Jose,CA, United States
Duration: 24 Jan 200126 Jan 2001

Other

OtherStorage and Retrieval for Media Databases 2001
CountryUnited States
CitySan Jose,CA
Period24/1/0126/1/01

Fingerprint

semantics
Semantics
retrieval
edge detection
Edge detection
Seed
seeds
Agglomeration
Color
color
interactions

Keywords

  • Clustering
  • Indexing
  • Object Extraction
  • Retrieval and Browsing
  • Tracking
  • Video Representation

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Fan, J., Yau, D. K. Y., Hacid, M. S., & Elmagarmid, A. (2001). Model-based semantic object extraction for content-based video representation and indexing. In M. M. Yeung, C. Li, & R. W. Lienhart (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 4315, pp. 523-535) https://doi.org/10.1117/12.410964

Model-based semantic object extraction for content-based video representation and indexing. / Fan, J.; Yau, D. K Y; Hacid, M. S.; Elmagarmid, Ahmed.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / M.M Yeung; C. Li; R.W. Lienhart. Vol. 4315 2001. p. 523-535.

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

Fan, J, Yau, DKY, Hacid, MS & Elmagarmid, A 2001, Model-based semantic object extraction for content-based video representation and indexing. in MM Yeung, C Li & RW Lienhart (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 4315, pp. 523-535, Storage and Retrieval for Media Databases 2001, San Jose,CA, United States, 24/1/01. https://doi.org/10.1117/12.410964
Fan J, Yau DKY, Hacid MS, Elmagarmid A. Model-based semantic object extraction for content-based video representation and indexing. In Yeung MM, Li C, Lienhart RW, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 4315. 2001. p. 523-535 https://doi.org/10.1117/12.410964
Fan, J. ; Yau, D. K Y ; Hacid, M. S. ; Elmagarmid, Ahmed. / Model-based semantic object extraction for content-based video representation and indexing. Proceedings of SPIE - The International Society for Optical Engineering. editor / M.M Yeung ; C. Li ; R.W. Lienhart. Vol. 4315 2001. pp. 523-535
@inproceedings{edc0184e1c944b54a1311d596aac4a36,
title = "Model-based semantic object extraction for content-based video representation and indexing",
abstract = "This paper proposes an integrated system for supporting content-based video retrieval and browsing over networks. An automatic semantic video object extraction technique for providing more compact video representation is developed. The video images are first partitioned into a set of homogeneous regions with accurate boundaries by integrating the results of color edge detection and region growing procedures. The object seeds, which are the intuitive and representative part of the semantic objects, are detected from these obtained homogeneous image regions. The semantic objects are then generated by a seeded region aggregation or a human interaction procedure. These obtained semantic objects are tracked along the time axis for exploiting their temporal correspondences among frames. Given the semantic video objects represented by a set of visual features, a seeded semantic video content clustering technique is developed for providing more effective video indexing, retrieval and browsing.",
keywords = "Clustering, Indexing, Object Extraction, Retrieval and Browsing, Tracking, Video Representation",
author = "J. Fan and Yau, {D. K Y} and Hacid, {M. S.} and Ahmed Elmagarmid",
year = "2001",
doi = "10.1117/12.410964",
language = "English",
volume = "4315",
pages = "523--535",
editor = "M.M Yeung and C. Li and R.W. Lienhart",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Model-based semantic object extraction for content-based video representation and indexing

AU - Fan, J.

AU - Yau, D. K Y

AU - Hacid, M. S.

AU - Elmagarmid, Ahmed

PY - 2001

Y1 - 2001

N2 - This paper proposes an integrated system for supporting content-based video retrieval and browsing over networks. An automatic semantic video object extraction technique for providing more compact video representation is developed. The video images are first partitioned into a set of homogeneous regions with accurate boundaries by integrating the results of color edge detection and region growing procedures. The object seeds, which are the intuitive and representative part of the semantic objects, are detected from these obtained homogeneous image regions. The semantic objects are then generated by a seeded region aggregation or a human interaction procedure. These obtained semantic objects are tracked along the time axis for exploiting their temporal correspondences among frames. Given the semantic video objects represented by a set of visual features, a seeded semantic video content clustering technique is developed for providing more effective video indexing, retrieval and browsing.

AB - This paper proposes an integrated system for supporting content-based video retrieval and browsing over networks. An automatic semantic video object extraction technique for providing more compact video representation is developed. The video images are first partitioned into a set of homogeneous regions with accurate boundaries by integrating the results of color edge detection and region growing procedures. The object seeds, which are the intuitive and representative part of the semantic objects, are detected from these obtained homogeneous image regions. The semantic objects are then generated by a seeded region aggregation or a human interaction procedure. These obtained semantic objects are tracked along the time axis for exploiting their temporal correspondences among frames. Given the semantic video objects represented by a set of visual features, a seeded semantic video content clustering technique is developed for providing more effective video indexing, retrieval and browsing.

KW - Clustering

KW - Indexing

KW - Object Extraction

KW - Retrieval and Browsing

KW - Tracking

KW - Video Representation

UR - http://www.scopus.com/inward/record.url?scp=0035054277&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035054277&partnerID=8YFLogxK

U2 - 10.1117/12.410964

DO - 10.1117/12.410964

M3 - Conference contribution

VL - 4315

SP - 523

EP - 535

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

A2 - Yeung, M.M

A2 - Li, C.

A2 - Lienhart, R.W.

ER -