Personality is the unique and stable way of thinking and behavior style that people have. Our study wants to explore the relationship between personality and behavior, especially gait. 179 graduate students participated in our research. First, all participants completed the Eysenck Personality Questionnaire (EPQ). EPQ divides personality into four dimensions: Extraversion/Introversion (E), Neuroticism/Stability (N), Psychoticism/Socialization (P) and Lie/Social Desirability (L). When the participants were asked walking for two minutes on a red carpet, we used Kinect to obtain the behavior data of the participants as the original data, and performed preprocessing such as denoising and coordinate translation on the original data. Then find the gait features that affect personality through regression analysis. We found that gait features can explain the N dimension to a very high degree, interpret the E and P dimensions to a relatively high degree, and interpret the L dimension to a moderate extent. The results show that gait can predict personality to a certain extent. Our study provides a new complement to the measurement of personality.