Executive Summary : | The project aims to develop an Artificial Intelligence-based algorithm that uses vision sensors to measure gaze and tissue motions around the eyes. The eye region, which includes the globe, eyelid margin, and skin, is highly expressive and conveys information about mental state, intentions, and emotions during human communication. The deformations of the globe and facial tissues cause the skin to deform in a characteristic way, forming wrinkles. These wrinkles depend on factors such as skin nature, age, sun damage, and facial deformities. The saccadic motions of the eye and surrounding skin are important indicators of mental attention, facial deformities, and other clinical disorders. Any irregularities in these rapid eye movements can infer psychopathologies, such as Progressive supranuclear palsy, Parkinson's disease, and Oculomotor apraxia. Other facial nerve disorders can manifest in abnormalities like Ptosis and abnormal blinking.
The project will propose an AI-based algorithm using deep architectures to track tissue movements and estimate parameters related to eye movements and tissue deformation. Experiments will use third-party data and custom vision-based face capture systems. The healthcare application, Globemetrics, will be developed based on the proposed algorithm for visualizing the deformation of periorbital tissue motion and related parameters. This will help in developing solutions for medical conditions, rapid health assessment, planning facial surgery, drowsiness detection, and patient-computer interaction systems. Considering the information periorbital tissue motions provide in understanding facial deformities, and the increasing availability of commodity camera devices, the project plans to explore video-based analysis of facial tissue motions in the eye region. |