Using generative probabilistic models for multimedia retrieval

T.H.W. Westerveld, Thijs Henk-Willem Westerveld

Research output: ThesisPhD Thesis - Research external, graduation UTAcademic

95 Downloads (Pure)

Abstract

This thesis discusses information retrieval from multimedia archives. Multimedia archives are collections of documents containing a mixture of text, images, video and audio. This work focuses on documents containing visual material and investigates search and retrieval in collections of images and video, where video is defined as a sequence of still images. No assumptions are made with respect to the content of the documents: the collections are not restricted to a specific domain (e.g., images of fingerprints or collections of x-ray pictures). Instead we concentrate on retrieval from generic, heterogeneous multimedia collections. In this research area a user¿s query typically consists of one or more example images and the implicit request is: ¿Find images similar to this one.¿ In addition the query may contain a textual description of the information need. The research presented here addresses three issues within this area.
Original languageUndefined
Supervisors/Advisors
  • de Jong, Franciska M.G., Supervisor
  • de Vries, A.P., Advisor
Thesis sponsors
Award date25 Nov 2004
Place of PublicationEnschede
Publisher
Print ISBNs90-75296-13-4
Publication statusPublished - 25 Nov 2004

Keywords

  • EWI-6572
  • METIS-220121
  • IR-41716

Cite this

Westerveld, T. H. W., & Westerveld, T. H-W. (2004). Using generative probabilistic models for multimedia retrieval. Enschede: Neslia Paniculata.
Westerveld, T.H.W. ; Westerveld, Thijs Henk-Willem. / Using generative probabilistic models for multimedia retrieval. Enschede : Neslia Paniculata, 2004. 192 p.
@phdthesis{a5fe62f856844eaa806f1b97db4f0cf2,
title = "Using generative probabilistic models for multimedia retrieval",
abstract = "This thesis discusses information retrieval from multimedia archives. Multimedia archives are collections of documents containing a mixture of text, images, video and audio. This work focuses on documents containing visual material and investigates search and retrieval in collections of images and video, where video is defined as a sequence of still images. No assumptions are made with respect to the content of the documents: the collections are not restricted to a specific domain (e.g., images of fingerprints or collections of x-ray pictures). Instead we concentrate on retrieval from generic, heterogeneous multimedia collections. In this research area a user¿s query typically consists of one or more example images and the implicit request is: ¿Find images similar to this one.¿ In addition the query may contain a textual description of the information need. The research presented here addresses three issues within this area.",
keywords = "EWI-6572, METIS-220121, IR-41716",
author = "T.H.W. Westerveld and Westerveld, {Thijs Henk-Willem}",
note = "CTIT Ph.D. Thesis Series No. 04-67",
year = "2004",
month = "11",
day = "25",
language = "Undefined",
isbn = "90-75296-13-4",
publisher = "Neslia Paniculata",

}

Westerveld, THW & Westerveld, TH-W 2004, 'Using generative probabilistic models for multimedia retrieval', Enschede.

Using generative probabilistic models for multimedia retrieval. / Westerveld, T.H.W.; Westerveld, Thijs Henk-Willem.

Enschede : Neslia Paniculata, 2004. 192 p.

Research output: ThesisPhD Thesis - Research external, graduation UTAcademic

TY - THES

T1 - Using generative probabilistic models for multimedia retrieval

AU - Westerveld, T.H.W.

AU - Westerveld, Thijs Henk-Willem

N1 - CTIT Ph.D. Thesis Series No. 04-67

PY - 2004/11/25

Y1 - 2004/11/25

N2 - This thesis discusses information retrieval from multimedia archives. Multimedia archives are collections of documents containing a mixture of text, images, video and audio. This work focuses on documents containing visual material and investigates search and retrieval in collections of images and video, where video is defined as a sequence of still images. No assumptions are made with respect to the content of the documents: the collections are not restricted to a specific domain (e.g., images of fingerprints or collections of x-ray pictures). Instead we concentrate on retrieval from generic, heterogeneous multimedia collections. In this research area a user¿s query typically consists of one or more example images and the implicit request is: ¿Find images similar to this one.¿ In addition the query may contain a textual description of the information need. The research presented here addresses three issues within this area.

AB - This thesis discusses information retrieval from multimedia archives. Multimedia archives are collections of documents containing a mixture of text, images, video and audio. This work focuses on documents containing visual material and investigates search and retrieval in collections of images and video, where video is defined as a sequence of still images. No assumptions are made with respect to the content of the documents: the collections are not restricted to a specific domain (e.g., images of fingerprints or collections of x-ray pictures). Instead we concentrate on retrieval from generic, heterogeneous multimedia collections. In this research area a user¿s query typically consists of one or more example images and the implicit request is: ¿Find images similar to this one.¿ In addition the query may contain a textual description of the information need. The research presented here addresses three issues within this area.

KW - EWI-6572

KW - METIS-220121

KW - IR-41716

M3 - PhD Thesis - Research external, graduation UT

SN - 90-75296-13-4

PB - Neslia Paniculata

CY - Enschede

ER -

Westerveld THW, Westerveld TH-W. Using generative probabilistic models for multimedia retrieval. Enschede: Neslia Paniculata, 2004. 192 p.