Katja Rus (2012) Smile detection. EngD thesis.
Facial expressions are one of the most powerful and indirect means of transferring feelings, cognitive status, ulterior motive and opinion among people. Considering their significance it is reasonable to expect that the development of tools recognizing facial expressions would have a big effect on everyday life. In the past years, the developers have made big progress on the development of classifiers that are needed for this kind of expression detection; however there is still a great gap between real life situations and laboratory experiments. This thesis is focused on researching the existing algorithms of smile detection, for which face detection must be executed beforehand. Various approaches, methods and databases that are being used will be presented and the development of two systems, proved to be most efficient, will be explained. In conclusion the functioning of smile detection on a simple computer program will be presented.
Actions (login required)