Title :
Modeling of Affective State Response to a Virtual 3D Face
Author :
Vidugiriene, Ausra ; Vaskevicius, Egidijus ; Kaminskas, Vytautas
Author_Institution :
Dept. of Syst. Anal., Vytautas Magnus Univ., Kaunas, Lithuania
Abstract :
Virtual environment is an important part of everyday life and it influences a psycho-emotional state of a user. This paper introduces affective state modeling using features of virtual 3D face. Observations of human affective state are done using preprocessed EEG (electroencephalogram) signals: excitement, meditation, frustration, engagement/boredom. The signals are measured using Emotiv EPOC device. Virtual 3D face with its feature modifications is used as a stimulus. A model that describes a relationship between stimulus and corresponding affective state signal is proposed. The unknown parameters of a model are estimated using auto- and cross-co variation functions and the method of least squares. Model validation showed that the model was able to predict the signals representing affective state of a human in a high accuracy.
Keywords :
electroencephalography; emotion recognition; face recognition; feature extraction; least squares approximations; psychology; virtual reality; EEG signals; Emotiv EPOC device; affective state modeling; affective state response modeling; auto-covariation functions; boredom state; cross-covariation functions; electroencephalogram signals; engagement state; excitement state; feature modifications; frustration state; least square method; meditation state; model validation; psychoemotional state; virtual 3D face features; virtual environment; Brain modeling; Electroencephalography; Face; Mathematical model; Nose; Predictive models; Three-dimensional displays; 3D face; affective state; dynamic model; model validation; parameter estimation; prediction;
Conference_Titel :
Modelling Symposium (EMS), 2013 European
Conference_Location :
Manchester
Print_ISBN :
978-1-4799-2577-3
DOI :
10.1109/EMS.2013.31
