bernsee.com > Face Recognition    PANDORA-V Face Recognition and Facial Expression Evaluation Technology We are the maker of the PANDORA-V parametric face recognition and facial expression evaluation technology. Development of this technology has begun in the 1990s and it has quickly become invaluable in analyzing the facial expression of a person to help in speech recognition ("lip reading"), face recognition (for speaker identification) and expression evaluation (audio-visual polygraph) in various projects at Prosoniq and the DSP Dimension. Beautification and Exposure Correction PANDORA-V is also successfully used in determining the correct exposure data for digital photography and for portrait beautification. It works by adapting a virtual wireframe model of a head and face to real video data and estimates the amount of contraction of the individual face muscles from the distortion matrix. Get Involved! PANDORA-V is an ongoing project and has seen many releases and improvements over the past 10 years. We are utilizing a 2 Terabyte database of reference data to train the artificial neural nets that drive the technology to detect a face and facial expression independent of cultural or ethnic background. We are constantly looking for additional face models to expand our database, so if you wish to participate please drop us an email. Sessions are being paid for by the hour and work involves counting numbers, reading predetermined text and dynamically interacting with a computer. Last change: 04.10.2006, ©2005-2006 S. M. Bernsee, all rights reserved. Content subject to change without notice. Content provided 'as is', see disclaimer.