The Centre For Digital Video Processing (CDVP) at Dublin City University, a research group with 7 Faculty members, 4 post-doctoral researchers and 20 postgraduate (M.Sc and Ph.D) students, prides itself on finding new and useful ways of working with Digital Media information. Its mission is to research, develop and evaluate techniques for the efficient management of large volumes of digital video information. Over the last 5 years many video retrieval systems have been developed allowing the searcher to effectively retrieval particular shots from huge collections of digital video. Please follow the above link to find out more details.

    "Information retrieval is the name of the process or method whereby a prospective user of the        information is able to convert his need for information into an actual list of citations to documents        in storage containing information useful to him"

This is one of the oldest definitions of information retrieval written by Mooers in the 1950's. With the development of many commercial information retrieval systems or search engines like Yahoo and Google in the 1990's, information retrieval has become well known to the majority of the population. The World Wide Web has become more accessible and useable, thanks to the facilities of these search engines. We are now able to navigate and browse through eight billion webpages using links alone.

Traditional information retrieval has operated over textual documents, with state of the art commercial information retrieval systems successfully searching and retrieving documents requested by the user. The development of new technologies integrating multiple media such as images, mp3s, audio and video have created vast libraries and archives in medical, criminal investigation, art galleries and TV broadcasters to name but a few. It is obvious there is a need/challenge for information management, organisation, retrieval and navigation through these vast multimedia archives. For example, the BBC archive stores an additional one million new items per year.

My main research concentrates on the content-based retrieval of digital video. A video information retrieval system is far more complex than a traditional information retrieval system dealing with textual data alone, as there are many different sources of evidences and information associated with images or video Although the detection of relevant documents is a fundamental requirement of a searcher during the course of the research we have identified problems which tend to hinder the searchers ability to interact with a system effectively, consequently we are not only looking for relevant documents covering a topic but are also interested in identifying whether or not these relevant documents contain new information about the topic. We assume a user has an information need and wants to gain knowledge about a topic. This we refer to as the detection of novel information from within a topic area or Novelty Detection.

Publications & Presentations

Dublin City University Video Track Experiments for TREC 2003 Browne P, Czirjek C, Gaughan G, Gurrin C, Jones G, Lee H Marlow S, McDonald K, Murphy N, O'Connor N, O'Hare N, Smeaton A, and Ye J. TRECVID 2003 - Text REtrieval Conference TRECVID Workshop, Gaithersburg, Maryland, 17-18 November 2003. (Full Pdf)

Design, Implementation and Testing of an Interactive Video Retrieval System
Gaughan G, Smeaton A, Gurrin C, Lee H and Mc Donald K. MIR 2003 - 5th ACM SIGMM International Workshop on Multimedia Information Retrieval, Berkeley, CA, 7 November 2003.(Full Pdf) Workshop homepage. [03-19] Presentation