Advancements in Emotion Recognition

Understanding the importance of human emotion in various aspects of life, the present paper seeks to contribute to the field of emotion recognition and identification using computer vision. Emotion detection is crucial in human-computer interaction since it offers a channel through which the interaction between human and computers can be made more natural. It is very important to understand the user’s emotions in various fields including marketing, human-centered computing, recommendation systems, and mental health monitoring. This survey is to further the development of more sophisticated and efficient technologies that are capable of identifying and perceiving the different aspects of emotions by dissecting the emotions of human beings. The traditional systems often limit the emotion detection to the seven basic emotions as the only possibility, thus not capturing the rich and diverse spectrum of human emotions. In order to overcome these limitations, the paper proposes to go a step further in this study by employing advanced features such as 3D models, local descriptors, and Convolutional Neural Networks (CNNs). In doing so, this not only enhance the ability of the system to identify the basic emotions but also come up with a way of differentiating the level of emotions overtime, thus adding another level of complexity and detail in the emotional perception. Also, the integration of temporal modelling to video emotion detection enables the identification of the temporal characteristics of emotions which can be useful in understanding how emotions evolve over time. This is because it has the potential of revolutionalizing human emotions’ understanding and application in the development of intelligent systems and thus influences the future development of intelligent systems’ relationship with human experiences.