AVideoDatabaseofMovingFacesandPeopleAliceJ.O’Toole,JoshuaHarms,.pdfVIP

812 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 27, NO. 5, MAY 2005 A Video Database of Moving Faces and People models of face and person recognition when the people (and their faces) are in motion and when the learning and test images/videos Alice J. O’Toole, Joshua Harms, are of different types. For example, one common application might Sarah L. Snow, Dawn R. Hurst, be to match a frontal mug shot of a face to a person walking past a Matthew R. Pappas, surveillance camera in a public place. Another common applica- Janet H. Ayyad, and tion is to match videos of people walking in different directions ´ Herve Abdi with respect to the camera. Motion tracking of the person, head, face, mouth/lips are also important prerequisite steps for many Abstract—We describe a database of static images and video clips of human recognition applications. faces and people that is useful for testing algorithms for face and person In addition to recognition-based applications, there are applica- recognition, head/eye tracking, and computer graphics modeling of natural human tions in human-computer interface design and computer graphics motions. For each person there are nine static “facial mug shots” and a series of for a database containing natural face and