A method for comparing content based image retrieval methods

Jacobus J Barnard, Nikhil V. Shirahatti

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

6 Citations (Scopus)

Abstract

We assume that the goal of content based image retrieval is to find images which are both semantically and visually relevant to users based on image descriptors. These descriptors are often provided by an example image - the query by example paradigm. In this work we develop a very simple method for evaluating such systems based on large collections of images with associated text. Examples of such collections include the Corel image collection, annotated museum collections, news photos with captions, and web images with associated text based on heuristic reasoning on the structure of typical web pages (such as used by Google(tm)). The advantage of using such data is that it is plentiful, and the method we propose can be automatically applied to hundreds of thousands of queries. However, it is critical that such a method be verified against human usage, and to do this we evaluate over 6000 query/result pairs. Our results strongly suggest that at least in the case of the Corel image collection, the automated measure is a good proxy for human evaluation. Importantly, our human evaluation data can be reused for the evaluation of any content based image retrieval system and/or the verification of additional proxy measures.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsS. Santini, R. Schettini
Pages1-8
Number of pages8
Volume5018
DOIs
StatePublished - 2003
EventInternet Imaging IV - Santa Clara, CA, United States
Duration: Jan 21 2003Jan 22 2003

Other

OtherInternet Imaging IV
CountryUnited States
CitySanta Clara, CA
Period1/21/031/22/03

Fingerprint

Image retrieval
retrieval
Museums
Websites
evaluation
museums
news

Keywords

  • Benchathlon
  • Benchmarking
  • Content based image retrieval
  • Image semantics
  • Performance

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Barnard, J. J., & Shirahatti, N. V. (2003). A method for comparing content based image retrieval methods. In S. Santini, & R. Schettini (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 5018, pp. 1-8) https://doi.org/10.1117/12.476179

A method for comparing content based image retrieval methods. / Barnard, Jacobus J; Shirahatti, Nikhil V.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / S. Santini; R. Schettini. Vol. 5018 2003. p. 1-8.

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

Barnard, JJ & Shirahatti, NV 2003, A method for comparing content based image retrieval methods. in S Santini & R Schettini (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 5018, pp. 1-8, Internet Imaging IV, Santa Clara, CA, United States, 1/21/03. https://doi.org/10.1117/12.476179
Barnard JJ, Shirahatti NV. A method for comparing content based image retrieval methods. In Santini S, Schettini R, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 5018. 2003. p. 1-8 https://doi.org/10.1117/12.476179
Barnard, Jacobus J ; Shirahatti, Nikhil V. / A method for comparing content based image retrieval methods. Proceedings of SPIE - The International Society for Optical Engineering. editor / S. Santini ; R. Schettini. Vol. 5018 2003. pp. 1-8
@inproceedings{9ec264a3a1f64f72919691abb8626e52,
title = "A method for comparing content based image retrieval methods",
abstract = "We assume that the goal of content based image retrieval is to find images which are both semantically and visually relevant to users based on image descriptors. These descriptors are often provided by an example image - the query by example paradigm. In this work we develop a very simple method for evaluating such systems based on large collections of images with associated text. Examples of such collections include the Corel image collection, annotated museum collections, news photos with captions, and web images with associated text based on heuristic reasoning on the structure of typical web pages (such as used by Google(tm)). The advantage of using such data is that it is plentiful, and the method we propose can be automatically applied to hundreds of thousands of queries. However, it is critical that such a method be verified against human usage, and to do this we evaluate over 6000 query/result pairs. Our results strongly suggest that at least in the case of the Corel image collection, the automated measure is a good proxy for human evaluation. Importantly, our human evaluation data can be reused for the evaluation of any content based image retrieval system and/or the verification of additional proxy measures.",
keywords = "Benchathlon, Benchmarking, Content based image retrieval, Image semantics, Performance",
author = "Barnard, {Jacobus J} and Shirahatti, {Nikhil V.}",
year = "2003",
doi = "10.1117/12.476179",
language = "English (US)",
volume = "5018",
pages = "1--8",
editor = "S. Santini and R. Schettini",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - A method for comparing content based image retrieval methods

AU - Barnard, Jacobus J

AU - Shirahatti, Nikhil V.

PY - 2003

Y1 - 2003

N2 - We assume that the goal of content based image retrieval is to find images which are both semantically and visually relevant to users based on image descriptors. These descriptors are often provided by an example image - the query by example paradigm. In this work we develop a very simple method for evaluating such systems based on large collections of images with associated text. Examples of such collections include the Corel image collection, annotated museum collections, news photos with captions, and web images with associated text based on heuristic reasoning on the structure of typical web pages (such as used by Google(tm)). The advantage of using such data is that it is plentiful, and the method we propose can be automatically applied to hundreds of thousands of queries. However, it is critical that such a method be verified against human usage, and to do this we evaluate over 6000 query/result pairs. Our results strongly suggest that at least in the case of the Corel image collection, the automated measure is a good proxy for human evaluation. Importantly, our human evaluation data can be reused for the evaluation of any content based image retrieval system and/or the verification of additional proxy measures.

AB - We assume that the goal of content based image retrieval is to find images which are both semantically and visually relevant to users based on image descriptors. These descriptors are often provided by an example image - the query by example paradigm. In this work we develop a very simple method for evaluating such systems based on large collections of images with associated text. Examples of such collections include the Corel image collection, annotated museum collections, news photos with captions, and web images with associated text based on heuristic reasoning on the structure of typical web pages (such as used by Google(tm)). The advantage of using such data is that it is plentiful, and the method we propose can be automatically applied to hundreds of thousands of queries. However, it is critical that such a method be verified against human usage, and to do this we evaluate over 6000 query/result pairs. Our results strongly suggest that at least in the case of the Corel image collection, the automated measure is a good proxy for human evaluation. Importantly, our human evaluation data can be reused for the evaluation of any content based image retrieval system and/or the verification of additional proxy measures.

KW - Benchathlon

KW - Benchmarking

KW - Content based image retrieval

KW - Image semantics

KW - Performance

UR - http://www.scopus.com/inward/record.url?scp=0037720323&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037720323&partnerID=8YFLogxK

U2 - 10.1117/12.476179

DO - 10.1117/12.476179

M3 - Conference contribution

AN - SCOPUS:0037720323

VL - 5018

SP - 1

EP - 8

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

A2 - Santini, S.

A2 - Schettini, R.

ER -