An impaired facial motor system is a common deficit associated with injury to the nervous system such as occurs during stroke or head trauma. Despite the impact Facial Motor System (FMS) damage has on psychological and social aspects of an individual’s quality of life, facial motor rehabilitation has received little attention until comparatively recently. In this paper preliminary results of our investigation into the use of machine vision methods for the development of an automatic feedback system are presented. We show by way of experimental results that our system provides initial steps towards the development of a system for enhancing the rehabilitation prospects of individuals suffering from FMS damage. The proposed system will act as an intelligent mirror, providing basic feedback by tracking key facial features during the attempted gesture. Such a system could potentially emulate the therapist, automatically assessing the patient according to standard facial disability measures, allowing a comprehensive record of patient engagement, performance, efficacy and outcome to be constructed in a relatively inexpensive manner.