We demonstrate a novel method of generating a mapping function which takes an image of a neutral face to an image of the same subject depicting an alternative expression. It is proposed that this mapping function can be used to automatically generate facial expressions from still images of never seen before faces. This technique draws on the work of Ekman's [8] Facial Action Coding System (FACS), which provides an anatomical basis for measuring facial movement. We use the FACS to generate a Facial Expression Texture Model (FETM), which is used in conjunction with several Artificial Neural Networks (ANN) to develop a mapping function. We describe this method in detail and provide results which demonstrate the effectiveness of this technique.