원문정보
초록
영어
Recently, due to the introduction of high-tech equipment in interactive exhibits, it became possible to measure various audiences’ responses in the interactive exhibition. Among them, this paper is to predict user’s emotion using the change of the facial features. In other words, the goal of this research is to build the interactive system with the emotion determination model to predict the audience’s response. At this time, as a technique to learn model to predict the response of the audience, we used an Artificial Neural Network-based model. And, as a model to present the emotion state of the audience, we used a Valence-Arousal model. In addition, this paper performed the experiment to compare the proposed model with the multiple regression-based model to see whether it had better performance or not. As a result, although the data set had much noise, we could confirm that the proposed model had better prediction accuracy.
목차
Introduction
Related Work
Valence-Arousal Model
Artificial Neural Network
Research Model
System architecture
The analysis technique of audience’s response: Artificia lNeural Network-based Model
Experiment design
Acquiring the audience’s data for analysis
Select the fields of dependent and independent variables
Extract significant variables
Design the experiment of ANN
Design the comparison model
Result
Conclusion
References