Home

Aspect Browser

Overview

AspectBrowser is a search interface which is designed to support vague, exploratory search tasks - the kinds of task where your not quite sure what your looking for, of how to describe what your looking for. The interface uses a storyboard-style metaphor, where you can create multiple aspects, each aspect allowing you to search for material, and also store any material you find in the tabbed area at the top of the interface. Using drag and drop, you can reorder results between aspects, and easily use the results of one search to start a new search - without touching any previous searches you've carried out. This is particularly useful when carrying out visual searches, where your search results are based on the visual similarity between examples you select, and the contents of a database. Such indexes of visual images or videos can therefore be searched without textual annotations or markup.

Features

There are a number of facilities and potential advantages to using the Aspect Browser

  • Fully automatic multimedia indexing functionality
  • Indexing and search of multiple different collections which can be a mixture of video or images
  • Flexible and novel search interface which supports multiple search aspects
  • Plug-in architecture for low-level image features

Demonstration

Click image for real-time demo. The Aspect Browser is optimised for use with Internet Explorer

Click here for a tutorial on how to use the system

Publications

  • Robert Villa, Joemon M. Jose, Hideo Joho and Ivan Cantador. An aspectual interface for supporting complex search tasks. In Proceedings of the 32nd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2009) (Accepted; In Press)
  • Robert Villa, Nicholas Gildea and Joemon M. Jose. FacetBrowser: a user interface for complex search tasks. In Proceedings of ACM Multimedia 2008, (ACM MM 2008), October 2008
  • Robert Villa, Nicholas Gildea and Joemon M. Jose. Collaborative awareness in multimedia search. In Proceedings of ACM Multimedia 2008 (ACM MM 2008), October 2008
  • Robert Villa, Nicholas Gildea and Joemon M. Jose. A study of awareness in multimedia search. In Proccedings of ACM/IEEE Joint Conference on Digital Libraries (JCDL 2008), June 2008
  • Thierry Urruty, Frank Hopfgartner, Robert Villa, Nicholas Gildea and Joemon M. Jose. A cluster-based simulation of facet-based search. In Proccedings of ACM/IEEE Joint Conference on Digital Libraries (JCDL 2008), June 2008
  • Robert Villa, Nicholas Gildea and Joemon M. Jose. A faceted interface for multimedia search. In Proceedings of the 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2008), July 2008