Dear Subscriber A Research Assistant position is available in the DIRC labs at Kingston University.
The position advert is attached to this email. Kind regards Dr Paolo Remagnino Digital Imaging Research Centre School of Computing and Information Systems Kingston University Penrhyn road Kingston upon Thames=20 Surrey, KT1 2EE telephone : +44 (0)20 8547 7930 facsimile: +44 (0)20 8547 7824 URL: www.king ------------------------------------------------------------------------------- Researcher Learning Computer Vision Systems Digital Imaging Research Centre School of Computing and Information Systems Following the successful first stage of the European-funded project ADVISOR (Annotated Digital Video for Intelligent Surveillance and Optimised Retrieval), a position as a Researcher is likely to be available from early 2002.The work is mainly concerned with the investigation and validation of learning approaches for the monitoring of pedestrians in public transport environments. These are part of a general effort to improve safety and security through the deployment of "intelligent" CCTV systems. This is a large project involving five other European partners and represents a major EU effort in distributed visual surveillance systems targeted at improving quality of life in public spaces. Demonstrators are planned in Brussels, Barcelona and London metropolitan railways systems. Direct involvement with public transport operators, manufacturers and system integrators provides an environment where the results from scientific and technological development can be tested in real scenarios. The post is expected to commence in February/March 2002 and to be available until the end of December 2002. The successful applicant is likely to be involved in semantic aspects of behaviour analysis. Other ADVISOR components provide image-derived information such as motion detection, tracking of people and basic event descriptors from which to build more complex behaviours. It is expected to involve specifically work in one or more of the following aspects related to learning behaviours of individuals and crowds important to operators (e.g. overcrowding, vandalism, entry into forbidden areas): * Identification, modelling and learning of behaviours: Investigation of modelling techniques such as Bayes nets, Markov models and stochastic grammars. * Multiple camera behaviour recognition. Two cases will be considered: (i) the simultaneous observation of a person by two cameras, and (ii) the passing of a person from the view of one camera to another. * Supervised learning. For the facilities necessary for feedback from trained human operators to be used to optimise the performance of the behaviour recognition and alarm generation software. This optimisation will be achieved by using supervised learning in the choice of algorithms and the adjustment of thresholds in the chosen algorithms in order to improve behaviour recognition, and reduce false alarm rates. The ideal candidate would have a background in Computer Science or similar subject, with a good publication record, proven ability to work as part of a multi-disciplinary team and good written and oral presentation skills. For further details please contact either Dr. Paolo Remagnino ([EMAIL PROTECTED]) or Dr. Sergio Velastin ([EMAIL PROTECTED]). The team is well known nationally and internationally for its work on visual surveillance. Salary will be up to 26,821 pa (depending on experience) including London Weighting.
