A New Social Emotion Estimating Method by Measuring Micro-movement of Human Bust

In this study, we propose a new micro-movement analyzing method of the human bust to estimate social emotions with human relations. In previous research for quantitatively measuring human intention and emotion, physiological sensors such as ECG, PPG, GSR, SKT, and EEG have been analyzed. However, these methods cannot avoid the measurement burden caused by sensor attachment. This may result in negative emotions that skew the true evaluation. To solve these problems, we focus on micro-body movement responding emotions. Micro-body movements are determined by analyzing successive image frames captured from a conventional webcam. The amount of the bust micro-movement is then obtained by subtracting two adjacent image frames. Based on the interval between the two image frames, the amounts of micro-movement per several frequency bands can be acquired. Because the calculated successive values of bust micro-movement are in the form of 1D temporal signal, it can be general method for all conventional temporal signal processing. The results showed less micro-movement in the case of an intimate relation group compared with the case of a non-intimate one.