RSS 1.0 Feed
RSS 2.0 Feed
Atom Feed
 

Description, analysis and evaluation of confidence estimation procedures for sub-categorisation

de Campos, T., Csurka, G., Perronnin, F., Hussain, Z., Shawe-Taylor, J., Pasupa, K., Saunders, C., Ali, H., Antenreiter, M., Ortner, R., Auer, P., Viitaniemi, V. and Laaksonen, J. (2009) Description, analysis and evaluation of confidence estimation procedures for sub-categorisation. Technical Report D6.2.1, Xerox Research Centre Europe, PinView.

Download

[img]
Preview
Published Version
PDF

3444Kb

Abstract

This report presents contributions in two main areas: the combination of low level image features with visual saliency maps and use of confidence measures for information fusion. For the first part, we show experiments with automatic saliency estimation methods based on bottom-up and top-down approaches. We also explored maps generated by mouse-clicks. These maps were used to give weights to local image features which are then used for image categorisation in an approach based on bag-of-patches. For the second part, we explored methods to associate confidence values to predictions of each test sample. These values are used to give per-sample weights for different information sources in classifier combination.

Creators:Teófilo de Campos, Gabriela Csurka, Florent Perronnin, Zakria Hussain, John Shawe-Taylor, Kitsuchart Pasupa, Craig Saunders, Haider Ali, Martin Antenreiter, Ronald Ortner, Peter Auer, Ville Viitaniemi, Jorma Laaksonen
Item Type:Technical Report
Research Group:Information - Signals, Images, Systems
Deposited On:15 Dec 2009 13:46 by Pasupa, Kitsuchart
Alternative Locations:http://www.pinview.eu/files/pinview-d6-2-1-final.p...
ID Code:18317
Last Modified:18 Feb 2010 16:26

Tools

Metadata

Download Statistics

Last month

Last year

Members of ECS may view the download statistics dashboard for this record.

Corrections

ECS staff and postgraduates may modify this record

  Welcome from Deputy Head of School (Research) Research Prospectus Industrial Partnerships New Research Students Notes for Guidance New Research Students Notes for Guidance
The ECS EPrints Repository supports OAI 2.0 with a base URL of http://eprints.ecs.soton.ac.uk/cgi/oai2

EPrints is free software developed by the University of Southampton to facilitate Open Access to research.
EPrints