AI-based Facial Recognition in Emotional Detections

Abstract

Facial recognition is a method of identifying a human face with the help of computer vision (CV). The popularity of smart gadgets and the advancement of cameras’ capabilities have caused facial recognition to become a hot topic among academia and practitioners. Besides the traditional facial data gathered from surveillance systems, commercial facial recognition systems for measuring emotional states have recently become popular. These systems are often AI-based and use facial recognition algorithms, along with biometrics, to map facial features and motions from an image or through a live stream. The aim of this paper is to study the ability of these systems to detect emotion accurately. Humans have complex personalities that are often present in our facial expressions and do not necessarily reflect specific feelings. Additionally, those with personality disorders, such as narcissistic or histrionic personality disorder, have different facial expressions than the general public. Those expressions are not representative of the emotions that would be detected through the diagnostic systems. Therefore, complementary technologies and solutions are needed to make such measurements more accurate.