Browse DORAS
Browse Theses
Latest Additions
Creative Commons License
Except where otherwise noted, content on this site is licensed for use under a:

Video retrieval using objects and ostensive relevance feedback

Browne, Paul (2005) Video retrieval using objects and ostensive relevance feedback. PhD thesis, Dublin City University.

Full text available as:

PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader


The thesis discusses and evaluates a model of video information retrieval that incorporates a variation of Relevance Feedback and facilitates object-based interaction and ranking. Video and image retrieval systems suffer from poor retrieval performance compared to text-based information retrieval systems and this is mainly due to the poor discrimination power of visual features that provide the search index. Relevance Feedback is an iterative approach where the user provides the system with relevant and non-relevant judgements of the results and the system re-ranks the results based on the user judgements. Relevance feedback for video retrieval can help overcome the poor discrimination power of the features with the user essentially pointing the system in the right direction based on their judgements. The ostensive relevance feedback approach discussed in this work weights user judgements based on the o r d e r in which they are made with newer judgements weighted higher than older judgements. The main aim of the thesis is to explore the benefit of ostensive relevance feedback for video retrieval with a secondary aim of exploring the effectiveness of object retrieval. A user experiment has been developed in which three video retrieval system variants are evaluated on a corpus of video content. The first system applies standard relevance feedback weighting while the second and third apply ostensive relevance feedback with variations in the decay weight. In order to evaluate effective object retrieval, animated video content provides the corpus content for the evaluation experiment as animated content offers the highest performance for object detection and extraction.

Item Type:Thesis (PhD)
Date of Award:2005
Supervisor(s):Smeaton, Alan F.
Uncontrolled Keywords:video retrieval; ostensive relevance feedback; decay weight
Subjects:Computer Science > Information retrieval
DCU Faculties and Centres:DCU Faculties and Schools > Faculty of Engineering and Computing > School of Computing
Use License:This item is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 3.0 License. View License
ID Code:17382
Deposited On:03 Sep 2012 14:21 by Fran Callaghan. Last Modified 20 Jan 2017 14:42

Download statistics

Archive Staff Only: edit this record