This book details a computational model of facial expression. This is a growing and relatively new type of problem within computer vision. One of the fundamental problems with previous techniques was the lack of a consistent method for measuring facial expression. The approach outlined in this book overcomes this through the use of an anatomical model of facial expression called the Facial Action Coding System (FACS). Using the FACS as a basis for describing expression we compute a Facial Expression Shape Model (FESM) and a Facial Expression Texture Model(FETM). These models are generated by statistically analysing a large dataset of facial images of individuals portraying speciﬁc expressions. The resulting low-dimensional models provide a framework for the representation, synthesis, and classiﬁcation of facial expression.