A Comparative Emotions-detection Review for Non-intrusive Vision-Based Facial Expression Recognition

Abstract

Affective computing advocates for the development of systems and devices that can recognize, interpret, process, and simulate human emotion. In computing, the field seeks to enhance the user experience by finding less intrusive automated solutions. However, initiatives in this area focus on solitary emotions that limit the scalability of the approaches. Further reviews conducted in this area have also focused on solitary emotions, presenting challenges to future researchers when adopting these recommendations. This review aims at highlighting gaps in the application areas of Facial Expression Recognition Techniques by conducting a comparative analysis of various emotion detection datasets, algorithms, and results provided in existing studies. The systematic review adopted the PRISMA model and analyzed eighty-three publications. Findings from the review show that different emotions call for different Facial Expression Recognition techniques, which should be analyzed when conducting Facial Expression Recognition.

Keywords: Facial Expression Recognition, Emotion Detection, Image Processing, Computer Vision