Recently, the use of the Internet has been largely influenced by increased participation and interaction of users. We are currently witnessing the mass-scale switchover among web content publishers from creating on-line content themselves to providing the tools and playgrounds for ordinary users to publish and share their self-produced multimedia content, such as photographs (Flickr.com) and video clips (YouTube.com). As a new web content production circle, user-generated multimedia content has enjoyed an enormous growth. For instance, YouTube videos account for 60% of all videos watched online in the UK and 45% in US. It has almost 20 million visitors to the site each month, according to Nielsen/NetRating. But users' ability to obtain useful information grows at a slow rate. Tools that support the effective recognition and retrieval of relevant information are still primitive.

This project aims at developing intelligent content-based indexing and retrieval tools for home users to annotate, search, and share their self-generated multimedia content (images/videos). To break down the difficulties raised by the so-called semantic gaps of multimedia content, we intend to create a new information indexing and seeking paradigm in which the hidden concept and relevancy of information items can automatically be determined by both considering the interactions or contributions of users (social-based) and analyzing the content of information items (content-based).

Supervisors
Jun Wang (Lecturer, University College London)
Ian Kegel (Head of Future Content Group, BT Group Research & Venturing)
Paul Marrow (Principal Multidisciplinary Researcher, Broadband Applications Research Centre, BT Group)

About the Candidate
The ideal candidate should have a strong background in machine learning (statistical pattern recognition), and also considerable programming experience. Some experience in multimedia content analysis is a plus. Funding is available for UK candidates only, or candidates that can demonstrate a relevant connection to the UK (usually established by being resident for a period of three years immediately before the date of application for an award). Detailed guidelines on eligibility for EPSRC CASE studentships are available on the EPSRC website. In addition, the student must have an EU citizenship to qualify for reduced UCL tuition fees.

About the Position
The PhD will be supervised by Jun Wang of University College London, and Ian Kegel and Paul Marrow of BT Research, located at Adastral Park. The position will be funded by EPSRC/BT Industrial CASE Award scheme (details can be found here). For eligible candidates, the funding will cover UCL tuition fees and provide a stipend of at least £15,300 per annum for 3 years.

Application and Information
If you wish to apply for this studentship, please use the link http://www.ucl.ac.uk/prospective-students/graduate-study/application-admission/. Make sure you specify PhD: "Multimedia Retrieval, Adastral Park Campus" on the part of the form that asks what you are applying for. Please ensure you mark the department as "Adastral Park" in your application and the name of the supervisor as Dr. Jun Wang.

Please also e-mail copy of your transcripts and a CV only to c.churcher 'at' adastral.ucl.ac.uk with the title "[PhD Studentship in Multimedia Retrieval - by Jun Wang]", or send by post to Mrs C Churcher, UCL Adastral Park Postgraduate Campus, Ross Building PP3, Adastral Park, Martlesham Heath, Suffolk, IP5 3RE, UK.

Informal enquiries may also be directed to Jun Wang, J (dot) Wang (at) adastral.ucl.ac.uk with the title "[PhD Studentship in Multimedia Retrieval]".

The closing date for applications is 31st December 2008.