2011 | OriginalPaper | Buchkapitel
A Highly Automated Method for Facial Expression Synthesis
verfasst von : Nikolaos Ersotelos, Feng Dong
Erschienen in: Transactions on Edutainment V
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
This paper proposes a highly automatic approach for a realistic facial expression synthesis that allows for enhanced performance in speed and quality, while minimizing user interferences. It will present a highly technical and automated method for facial feature detection, by allowing users to perform their desired facial expression synthesis with very limited labour input. Moreover, it will present a novel approach for normalizing the illumination settings values between the source and the target images, thereby allowing the algorithm to work accurately, even in different lighting conditions. We will present the results obtained from the proposed techniques, together with our conclusions, at the end of the paper.