A Framework for Multimodal Affect recognition Using Physiological Signals

Affect recognition is becoming an important field for human computer interaction. Recently, there has been a significant amount of work on affect recognition using facial expression and speech analysis. However, the influence of affects should also effectively reflect in physiological signals providing substantial advantage of being robust against possible artifacts of human social marking.  This research introduces physiological signals for affect recognition along with other modalities such as text, speech and visual analysis (multimodal). The  framework is targeted for learning systems but will be flexible to other applications too.

Supervisor: Dr. Rafael A. Calvo at Electrical and Information Engineering, University of Sydney

Co-supervisor: Dr. André van Schaik at Electrical and Information Engineering, University of Sydney