Realtime Action Unit Detection
The automatic detection of emotions in faces plays an increasing role in todays life. The goal of this work is to provide a simple and flexible Framework for Action Unit detection with Support Vector Machines. Therefore, I present the project for which this framework was made, a short overview of the theories used, the database used for training, the created framework, including landmark estimation and normalization, feature extraction, training Support Vector Machines and predicting Action Units with them.
Bachelor of Science (B.Sc.)