Abstract:
© 2019, Springer Nature Switzerland AG. In this paper, we explore the possibility of using the visual and sound stimuli obtained in various incidents when immersed in virtual reality, to detect human emotion by measuring the human bio-signals: heart rate, electroencephalogram (EEG), blood volume pressure, skin temperature and galvanic skin response (GSR) using bio-sensors. Further classification of signals occurs using a neural network. The received statistical characteristics are used as a contribution to the neural network for classification according to the Lövheim cube of emotions. The resulting algorithm for recognizing emotions based on human bio-signals in virtual reality will be used to predict emotional reactions to various events in virtual environments and, consequently, to increase their immersion.