CLEF logo
ImageCLEF 2004

Text and/or Content-Based Cross Language Image Retrieval


Introduction

To promote research in cross language image retrieval we instigated the ImageCLEF campaign as part of the Cross Language Evaluation Forum (CLEF). We felt this contribution to the CLEF campaign would address an important and timely problem not dealt with by existing cross language evaluations. We envisage ImageCLEF will appeal to both commercial and academic research communities including: cross language information retrieval, image retrieval, and user interaction. The main aims of the ImageCLEF campaign are:
  • To promote and initiate international research for CL image retrieval.
  • To further our understanding of the relationships between CL texts and images for IR.
  • To create a set of useful standardised resources for CL image retrieval to scientific communities in the whole.
ImageCLEF consists of three different tasks based on different domains, scenarios and collections. The campaign is run in a similar manner as the TREC and NTCIR information retrieval evaluations and we encourage participants to take part in one (or all) of the following system and user-evaluation tasks:
  • A bilingual ad hoc retrieval task in which multilingual search requests (topics) are to be performed against an historic photographic collection in which images are accompanied with English captions.
  • A medical retrieval task in which an example image is used to perform a search against a medical image database consisting of images such as scans and x-rays to find similar images. Each medical image or a group of images represents an illness, and case notes in English or French are associated with each illness to be used for diagnosis.
A preliminary flyer for ImageCLEF 2004 can be downloaded here and a more detailed plan of the tasks offered by the campaign can be found here.

I'm interested so how do I register?
You can register for ImageCLEF 2004 by contacting Carol Peters, the main coordinator for CLEF. For more specific information about any aspect of ImageCLEF or the tasks, please contact Paul Clough.

Once you have registered your interest with Carol Peters and filled in the appropriate copyright and declaration forms you will be able to download the data collections. The data can be downloaded via our ftp server, but if you experience any difficulties or you would rather access the collection via another method (e.g. CDROM) then please let Paul Clough know to arrange this.

Data Collections
There are two collections available for ImageCLEF 2004:
  1. the St Andrews photographic archive which is to be used for both the ad hoc and the interactive tasks [More Information] [Download ~ 723MB].
  2. the CasImage medical database provided by University Hospitals Geneva [More Information] [Download ~ 1.3GB].

Click on the above links to find out more information about each collection and how these can be obtained.

Important dates
   
Registration opens 15 January 2004
Data release from February 2004
Topics release from 15 March 2004
Submission of runs 31 May 2004
Release of results from 15 July 2004
Workshop papers deadline 15 August 2004
Workshop (in conjunction with ECDL in Bath, UK) 15-16 September 2004
   

Results

Results for ImageCLEF 2004 can be found here:B An overview paper was written about ImageCLEF for the CLEF workshop and the CLEF proceedings (Lecture Notes in Computer Science) summarising the evaluation campaign.

Organisers of this track

Paul Clough, University of Sheffield (p.d.clough@sheffield.ac.uk)
Mark Sanderson, University of Sheffield (m.sanderson@sheffield.ac.uk)
Henning Mueller, University Hospitals of Geneva (henning.mueller@sim.hcuge.ch)

We have set up a mailing list: imageclef@sheffield.ac.uk for participants. Please contact Paul Clough to be added to the list.


Last Modified: January 2004 By: Paul Clough