Facing the reality of semantic image retrieval

P.G.B. Enser, C.J. Sandom, J. Hare, P.H. Lewis

Research output: Contribution to journalArticle

Abstract

Purpose – To provide a better-informed view of the extent of the semantic gap in image retrieval, and the limited potential for bridging it offered by current semantic image retrieval techniques. Design/methodology/approach – Within an ongoing project, a broad spectrum of operational image retrieval activity has been surveyed, and, from a number of collaborating institutions, a test collection assembled which comprises user requests, the images selected in response to those requests, and their associated metadata. This has provided the evidence base upon which to make informed observations on the efficacy of cutting-edge automatic annotation techniques which seek to integrate the text-based and content-based image retrieval paradigms. Findings – Evidence from the real-world practice of image retrieval highlights the existence of a generic-specific continuum of object identification, and the incidence of temporal, spatial, significance and abstract concept facets, manifest in textual indexing and real-query scenarios but often having no directly visible presence in an image. These factors combine to limit the functionality of current semantic image retrieval techniques, which interpret only visible features at the generic extremity of the generic-specific continuum. Research limitations/implications – The project is concerned with the traditional image retrieval environment in which retrieval transactions are conducted on still images which form part of managed collections. The possibilities offered by ontological support for adding functionality to automatic annotation techniques are considered. Originality/value – The paper offers fresh insights into the challenge of migrating content-based image retrieval from the laboratory to the operational environment, informed by newly-assembled, comprehensive, live data.
Original languageEnglish
Pages (from-to)465-481
Number of pages17
JournalJournal of Documentation
Volume63
Issue number4
DOIs
Publication statusPublished - Jul 2007

Fingerprint

Image retrieval
Functionality
Annotation
Query
Indexing
Factors
Test collections
Design methodology
Efficacy
Scenarios
Metadata
Paradigm

Keywords

  • Image retrieval, Semantics

Cite this

Enser, P. G. B., Sandom, C. J., Hare, J., & Lewis, P. H. (2007). Facing the reality of semantic image retrieval. Journal of Documentation, 63(4), 465-481. https://doi.org/10.1108/00220410710758977
Enser, P.G.B. ; Sandom, C.J. ; Hare, J. ; Lewis, P.H. / Facing the reality of semantic image retrieval. In: Journal of Documentation. 2007 ; Vol. 63, No. 4. pp. 465-481.
@article{6eec9b3af8de4d748a37975dae92e589,
title = "Facing the reality of semantic image retrieval",
abstract = "Purpose – To provide a better-informed view of the extent of the semantic gap in image retrieval, and the limited potential for bridging it offered by current semantic image retrieval techniques. Design/methodology/approach – Within an ongoing project, a broad spectrum of operational image retrieval activity has been surveyed, and, from a number of collaborating institutions, a test collection assembled which comprises user requests, the images selected in response to those requests, and their associated metadata. This has provided the evidence base upon which to make informed observations on the efficacy of cutting-edge automatic annotation techniques which seek to integrate the text-based and content-based image retrieval paradigms. Findings – Evidence from the real-world practice of image retrieval highlights the existence of a generic-specific continuum of object identification, and the incidence of temporal, spatial, significance and abstract concept facets, manifest in textual indexing and real-query scenarios but often having no directly visible presence in an image. These factors combine to limit the functionality of current semantic image retrieval techniques, which interpret only visible features at the generic extremity of the generic-specific continuum. Research limitations/implications – The project is concerned with the traditional image retrieval environment in which retrieval transactions are conducted on still images which form part of managed collections. The possibilities offered by ontological support for adding functionality to automatic annotation techniques are considered. Originality/value – The paper offers fresh insights into the challenge of migrating content-based image retrieval from the laboratory to the operational environment, informed by newly-assembled, comprehensive, live data.",
keywords = "Image retrieval, Semantics",
author = "P.G.B. Enser and C.J. Sandom and J. Hare and P.H. Lewis",
year = "2007",
month = "7",
doi = "10.1108/00220410710758977",
language = "English",
volume = "63",
pages = "465--481",
journal = "Journal of Documentation",
issn = "0022-0418",
number = "4",

}

Enser, PGB, Sandom, CJ, Hare, J & Lewis, PH 2007, 'Facing the reality of semantic image retrieval', Journal of Documentation, vol. 63, no. 4, pp. 465-481. https://doi.org/10.1108/00220410710758977

Facing the reality of semantic image retrieval. / Enser, P.G.B.; Sandom, C.J.; Hare, J.; Lewis, P.H.

In: Journal of Documentation, Vol. 63, No. 4, 07.2007, p. 465-481.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Facing the reality of semantic image retrieval

AU - Enser, P.G.B.

AU - Sandom, C.J.

AU - Hare, J.

AU - Lewis, P.H.

PY - 2007/7

Y1 - 2007/7

N2 - Purpose – To provide a better-informed view of the extent of the semantic gap in image retrieval, and the limited potential for bridging it offered by current semantic image retrieval techniques. Design/methodology/approach – Within an ongoing project, a broad spectrum of operational image retrieval activity has been surveyed, and, from a number of collaborating institutions, a test collection assembled which comprises user requests, the images selected in response to those requests, and their associated metadata. This has provided the evidence base upon which to make informed observations on the efficacy of cutting-edge automatic annotation techniques which seek to integrate the text-based and content-based image retrieval paradigms. Findings – Evidence from the real-world practice of image retrieval highlights the existence of a generic-specific continuum of object identification, and the incidence of temporal, spatial, significance and abstract concept facets, manifest in textual indexing and real-query scenarios but often having no directly visible presence in an image. These factors combine to limit the functionality of current semantic image retrieval techniques, which interpret only visible features at the generic extremity of the generic-specific continuum. Research limitations/implications – The project is concerned with the traditional image retrieval environment in which retrieval transactions are conducted on still images which form part of managed collections. The possibilities offered by ontological support for adding functionality to automatic annotation techniques are considered. Originality/value – The paper offers fresh insights into the challenge of migrating content-based image retrieval from the laboratory to the operational environment, informed by newly-assembled, comprehensive, live data.

AB - Purpose – To provide a better-informed view of the extent of the semantic gap in image retrieval, and the limited potential for bridging it offered by current semantic image retrieval techniques. Design/methodology/approach – Within an ongoing project, a broad spectrum of operational image retrieval activity has been surveyed, and, from a number of collaborating institutions, a test collection assembled which comprises user requests, the images selected in response to those requests, and their associated metadata. This has provided the evidence base upon which to make informed observations on the efficacy of cutting-edge automatic annotation techniques which seek to integrate the text-based and content-based image retrieval paradigms. Findings – Evidence from the real-world practice of image retrieval highlights the existence of a generic-specific continuum of object identification, and the incidence of temporal, spatial, significance and abstract concept facets, manifest in textual indexing and real-query scenarios but often having no directly visible presence in an image. These factors combine to limit the functionality of current semantic image retrieval techniques, which interpret only visible features at the generic extremity of the generic-specific continuum. Research limitations/implications – The project is concerned with the traditional image retrieval environment in which retrieval transactions are conducted on still images which form part of managed collections. The possibilities offered by ontological support for adding functionality to automatic annotation techniques are considered. Originality/value – The paper offers fresh insights into the challenge of migrating content-based image retrieval from the laboratory to the operational environment, informed by newly-assembled, comprehensive, live data.

KW - Image retrieval, Semantics

U2 - 10.1108/00220410710758977

DO - 10.1108/00220410710758977

M3 - Article

VL - 63

SP - 465

EP - 481

JO - Journal of Documentation

JF - Journal of Documentation

SN - 0022-0418

IS - 4

ER -