Keynote Speaker
Tim Cootes
Title: Deformable Object Modelling and Matching
Abstract
Statistical models of the shape and appearance of deformable objects have become widely used in Computer Vision and Medical Image Analysis. In this talk I will give an overview of statistical shape and appearance modelling and of a range of efficient algorithms for matching such models to new images (Active Appearance Models and their variants). I will also describe recent work on automatically constructing models from minimally labelled training images. Examples will be shown both from facial image analysis and medical image interpretation fields.
Biography
Tim Cootes recieved his degree in Maths and Physics, then completed a PhD in Civil Engineering, before moving into the field of Computer Vision in 1991.
He became a lecturer at the University of Manchester in 2001, and a Professor of Computer Vision in 2006.
His work has focused on building statistical models of deformable objects, and algorithms for matching such models to new images. He was involved in the development of the widely used Active Shape Model and Active Appearance Models, and more recently has been developing methods for group-wise registration of shapes and images, which allow automatic construction of such models. He is particularly interested in facial image analysis, and the interpretation of medical images.
Tim's
homepage
