Thesis supervisor: Cecília Sikné Lányi
Location of studies (in Hungarian): University of Pannonia, H-8200 Veszprém, Egyetem u. 10. Hungary Abbreviation of location of studies: PE
Description of the research topic:
Emotions and expressions by virtual agents are two important issues for future human-machine interfaces. In fact, recent development in the psychology of emotions and the progression in graphic computing techniques allow us today to animate virtual personages realistically and enable them to express themselves.
Facial animation is one of the most important parts in the field of virtual agents.
Facial animation transfer aims to animate a target face (virtual) from data extracted from a source face (virtual or real, a real image or video).
The main goal of the research is to improve the process of creating facial animations, making them more realistic, and then using these animations in real-time applications. This will allow the application to respond to the user's emotions.
To achieve this goal the research plan is divided into five stages:
1. Studying and understanding different existent methods of real-time facial animation transfers.
2. Creating target models (faces and virtual avatars) and automatically detect characteristic points of the source face.
3. Recognition of user emotion.
4. Designing and implementing a real-time facial animation tool.
5. Validating the whole system.
The output of the research is to develop algorithms to help IT engineers develop better and faster, real-time face animations with emotions. One potential target group could be people with stroke. Avatars with emotions could help their rehabilitation.
Papers related to the topic:
Lei Zhang, Yang Wang, Sen Wang, Dimitris Samaras, Image-Driven Re-targeting and Relighting of Facial Expressions In Computer Graphics International, pp. 11-18, (2005).
Benkő A., Sik Lányi C., Stark J.: Interacting via a Virtual Language Therapy Room, Computer-Based Intervention and Diagnostic Procedures - Applications for Language-Impaired Persons Workshop, 7-8 July 2008, Vienna, Austria, pp. 19-2
Benkő A., Sik Lányi C.: Ambient Intelligence in Multimedia and Virtual Reality Environments for the rehabilitation, Multimedia Techniques For Device and Ambient Intelligence by Damiani E. and Jeong J (Eds.) Springer-Verlag, 2009, pp. 165-178.
Kai Li, Qionghai Dai, Ruiping Wang Yebin Liu, A Data-Driven Approach for Facial Expression Retargeting in Video, IEEE Transactions on Multimedia 16(2):299-310 (2014)
Required language skills: english Number of students who can be accepted: 1
Deadline for application: 2021-02-28
2024. IV. 17. ODT ülés Az ODT következő ülésére 2024. június 14-én, pénteken 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).
Login
