Luana Bezerra Batista, Herman Martins Gomes
Universidade Federal de
Campina Grande
Departamento de Sistemas e Computação
Abstract. Facial
Expression Recognition Systems (FERS) are usually applied to human-machine
interfaces, enabling the utilization of services that require a good
identification of the emotional state of the user. This work presents a new
view of the facial expression recognition problem, by addressing the question
of whether or not is possible to classify previously labeled photogenic and non-photogenic
face images, based on their appearance. In the proposed approach, a Support
Vector Machine (SVM) is trained with Gabor
representations of the face images - extracted from their left-side - to learn
the relationships between facial expressions and the concept of a good
photography of the face of a person. The obtained results showed that Gabor filters and SVM are promising, having presented good
recognition rates.
[ contact authors ]
Proposed Approach |
Examples of photogenic and non-photogenic pictures from Cohn-Kanade Database |