Ανάπτυξη εφαρμογής με εικόνες συναισθηματικής αλληλεπίδρασης

Μπουγιούκου, Βικτωρία

Application development eith images of emotional interaction

Master Thesis

Author

Μπουγιούκου, Βικτωρία

Date

2014-06-17

Abstract

Due to the growing study of the human emotions in recent years, is perceived the great importance in many scientific fields. Many theories of emotions have developed some of them consider the emotions distinct states and others place them on pillars, emphasizing the association that exists between them. The major researcher of emotions is Ekman, according to him there are six basic emotions which occur all people and expressed with the same facial expressions. However, the theory of emotions could not miss from Computer Science and particularly from the human-machine communication. So, a new field has created, Affective Computing. The development of future systems, capable to understand human affective states and respond to emotional changes is the key aim of Affective Computing. In this context, Automatic Affect Recognition is a major challenge, yet at the same time, a significantly complex problem. Solutions to this problem can lead to advances in human-computer interaction, and may as well lead to improvement of the quality of life. The present work addresses issues defining, understanding and representation of emotional episodes from the standpoint of theoretical sciences and especially psychology, discusses some of the basic techniques used by the computer to retrieve information about the user's emotional state through facial expressions and approaches the subject of animation of virtual characters with special emphasis on synthetic faces of MPEG-4 and motion parameters of the face, called FAPs (Facial Animation Parameters). Finally, there is a description of Delphi developed application, which concerns the understanding and recognition of basic emotions of children on the autism spectrum.