COMS 4701 Artificial Intelligence

Facial Recognition remains one of the most popular fields in computer imaging research due to its broad scope of applications in identity verification, human computer interactions, law enforcement applications and video surveillance systems. Although numerous artificial intelligence and machine learning techniques have been successfully applied to still image recognitions, recognition in video streams remains a difficult task due to the changing conditions present in the video data, such as lighting, pose, and image motion. This paper describes an application to solve the above problem with automatic image selection and normalization from video streams using OpenCV, and conducting recognition through training and testing using Machine Learning library WEKA.