Lie Detection by Facial Expressions in Real Time

Talking is a way of human communication in everyday life. During a conversation, your partner sometimes lies to you, and you cannot even know that. For this reason, there have been many in-depth studies on lie detection. According to some latest scientific studies, liars will have a few facial expressions such as eye blinking, lips retracted, lips corners raising, head movement, eyebrows frowning, or eyebrows raising. These may not be similar due to the facial features of each person, different living habits, but these differences are insignificant. After synthesizing features from these studies, we based on some computer vision and signal processing algorithms to design a real-time algorithm that can score and predict whether an opponent is lying or not. Our goal is for the algorithm to recognize the liars after a short conversation by the laptop camera. Besides, our algorithm can also deploy to other high-end hardware devices to improve the accuracy of the prediction. Finally, the results can be inaccurate because the speaker often knows how to hide these features in real life. However, we know that the results are incompletely believable, but they partly reflect the state of the subject, and you can make decisions based on that.