SEDUCTION (Project Idex)
2014-2015 : This project aims to ananlyse the variation of facial expressions in the multicultural environment using combined acoustic and visual information with the machine learning methods:
- Collabration labotories: LaBRI, University of Kyoto
- with Aurélie Bugeau, Jean-Luc ROUAS, Takaaki SHOCHI, Toyoaki Nishida
Reasearch methods:
Key words: facial experssion, Action units, multi-kenerl SVMs
In this work, we propose a method using the fusion of the different appearance and geometry features based on a multi-kernel Support Vector Machine (SVM) for the automatic estimation of the variation of facial expression in terms of intensities of Action Units (AUs). The result of our approach benefiting from taking advantages of the different features adapting to a multi-kernel SVM is shown to outperform the conventional methods based on the mono-type feature with single kernel SVM.