CVPR at the University of York

Computer Vision and Pattern Recognition at the University of York

Our research spans a wide range of topics, from theoretical aspects of pattern recognition to the practical application of computer vision. The overall philosophy of the group is to bring the objective principles of pattern recognition to the design of robust and effective algorithms for machine vision.

Some of fundamental questions being asked are:

How can representations of visual information automatically adapt to changing environments?
How can relational models be matched most effectively against data that is highly corrupted?
How can the different levels of representation interact most effectively in a vision system when observational uncertainty is a limiting factor?

The mathematical framework for this is provided by information theory (especially Bayesian methods), statistical physics and optimisation theory. Vision tasks under study include face recognition and modelling, polarisation imaging, reflectance modelling, diffusion tensor imaging, shape-from-shading and stereo. We also carry out research on relational and graph descriptions of patterns, including matching, partitioning, embedding, clustering and generation. For more details of our research, see our research pages

The CVPR Icon

This picture arose out of trying to debug a procedure for computing constrained Delaunay triangulations, and is not particularly informative. We have adopted it as an icon since it underlines some of the complexities in seemingly very simple vision tasks.