Facial Expression Recognition

At VicarVision we develop FaceReader, world’s first commercial tool capable of automatically analyzing facial expressions, providing users with an objective assessment of a person’s emotion.

FaceReader uses an advanced 3D face modeling technique, with over 500 keypoints. The system is capable of modeling a face in realtime, without any manual initialization needed.

faceModeling

In addition to facial expressions, we develop and research state-of-the-art computer vision and machine learning concepts. Our related publications are found below:

[2014] H. Emrah Tasli, Paul Ivan; “Turkish Presidential Elections TRT Publicity Speech Facial Expression Analysis”; arXiv:1408.3573

[2015] Amogh Gudi, H. Emrah Tasli, Tim M. den Uyl, Andreas Maroulis; Deep Learning based FACS Action Unit Occurrence and Intensity Estimation, Int. Conf. on Automatic Face and Gesture Recognition, Facial Expression Recognition and Analysis Challenge FERA 2015.

[2015] Tim M. den Uyl, H. Emrah Tasli, Paul Ivan, Mariska Snijdewind; Who do you want to be? Real-time Face Swap; Demo Paper at International Conference on Automatic Face and Gesture Recognition, FG2015.

[2015] H. Emrah Tasli, Tim M. den Uyl, Hugo Boujut, Titus Zaharia; Real-Time Facial Character Animation; Demo Paper at International Conference on Automatic Face and Gesture Recognition, FG2015.

[2005] Marten J. den Uyl, Hans van Kuilenburg. The FaceReader: Online facial expression recognition. Proceedings of Measuring Behaviour; 5th International Conference on Methods and Techniques in Behavioural Research.

homepage