Computational Modelling Group

Seminar  16th November 2011 5 p.m.  B53/4025

[IEEE] Semantic Data Fusion in Biometrics: Augmenting Recognition Capability using Human Descriptions

Professor Mark Nixon
University of Southampton

Categories
Biometrics, Computer Science, Scientific Computing, Software Engineering
Submitter
Camillia Zedan

Soft biometrics is an emerging area of interest in biometrics: can we augment computer vision derived measures by human descriptions and if so, what is the interrelationship between them. We have been developing new approaches in gait biometrics, based on human descriptions which can extend the descriptions derived by automatic means, say by using computer vision. The human descriptions are semantic and are a set of labels which are converted into numbers. Naturally, there are considerations of language and psychology when the labels are collected. This talk will describe how the labels are collected, how they can be used to enhance recognising people by the way they walk (we also use the approaches to recognise vehicles by sound). A new development of this approach might lead to a new procedure for collecting witness statements. An earlier version of this talk was originally an invited plenary talk within the IEEE International BiDs conference at the Biometric Consortium Conference 2009 http://www.biometrics.org/bc2009/, in Tampa FL.