This intensive course is aimed at postgraduate research students in the areas of Computer Vision, Image Processing and Pattern Recognition. It is sponsored by the Engineering and Physical Sciences Research Council (EPSRC) and organised with the assistance of the British Machine Vision Association and Society for Pattern Recognition (BMVA).
The course is residential, spanning five days (midday Monday to midday Friday), and consists of lectures, tutorials and practicals on diverse topics in computer vision, image processing and pattern recognition. It is intended that the course will complement existing technical lecture course material that most students meet in the first year of their postgraduate training. It will provide an opportunity to broaden awareness of available vision, image and pattern recognition techniques and to develop skills in research methodology. The course will mix state of the art presentations from acknowledged UK experts with group work, case studies and practical exercises.
For EPSRC sponsored research students the course is FREE . The registration fee for non-EPSRC students will be £500. A limited number of bursaries worth £250 will be available from the BMVA to assist non-EPSRC students to attend the course.
The number of places is limited and therefore participants will be selected on the basis of their experience and suitability to benefit from the course. Priority will be given to students in the first or second year of their postgraduate study. Applicants must provide brief details of their research area and a statement of the benefit of the course to their research. An endorsement is required from the student's research supervisor.
Applicants will be advised of the fate of their application as soon as possible but by mid May at the latest.
Biological vision systems. Image acquisition and image modelling. Filtering, feature extraction and perceptual grouping. Statistical pattern recognition and decision making methods. 3D object recognition and reconstruction. Motion analysis. Theories of high-level vision.
The art of literature review. Ad-hoc techniques versus a framework-based methodology. Model development. Experiment design. Statistical analysis. Robust estimation. Performance characterisation. Research work reporting and presentation.
Planning: project formulation, research problem identification, definition of research objectives. Work-programme definition, methods to be used. Experimental evaluation. Work-programme schedule and monitoring. Programme revision. Cooperation and collaboration.
Illustration of methodology development on the problem of edge detector design. Medical/industrial application case studies.
Image processing environments. Image processing algorithm libraries. Public domain software. Software sharing, and ethics. Software tools. Standardisation. Visualisation. System/software integration.
The school programme will offer ample opportunity for the active involvement of the participants in group activities, personal research project presentation and experimentation using Unix based workstations.