| Abstract |
Facial Emotion Recognition (FER) plays a pivotal role in the realm of Smart Education Systems (SES), catering to the growing need for personalized and adaptive learning experiences. This paper presents a comprehensive review of the current state-of-the-art techniques and methodologies employed in FER in general and within the context of SES in particular. The primary objective of this review is to explore the FER technology for the enhancement of educational environments. The review begins by providing an insightful overview of the significance of emotions in education, leveraging deep learning approaches to elucidate the impact of emotional states on cognitive performance and academic outcomes. Subsequently, it explores the key components and working principles of FER systems, encompassing the extraction, representation, and classification of facial emotions. Furthermore, this paper highlights the updated literature in the field of FER applied to SES, ranging from real-time emotion analysis during classroom interactions to adapting instructional content based on learners' emotional states. Furthermore, the review critically analyzes the existing facial emotion recognition datasets utilized in facial emotion recognition research. The dataset is subcategorized in to two parts, images and videos, providing readers with convenience in understanding the various data sources employed in FER model training and evaluation. © 2023 IEEE. |
