Skip to main content

Unconstrained Face Recognition using Cell Phone Devices: Faces in the Wild Thirimachos Bourlai of West Virginia University and Dimitris Metaxas of Rutgers University


A new cell phone face database is necessary for studying the problem of unconstrained face recognition. Information collection for the database centers on obtaining facial data generated by volunteer project participants with an ideal total of 200 successfully scanned. This project involves matching the probe face images captured at variable distances by different portable devices (such as cell phones), against the good-quality face images (e.g., mug shots) acquired within rigorously controlled conditions (indoors, with fixed standoff distance between the subject and the camera, cooperative subjects, controlled illumination) using the same cell phones. The database is collecting facial data with different phones under different conditions in order to improve facial recognition technology.