FAKE EMOTION DETECTION

One of the most important topics in emotion detection is the ability to differentiate between real and fake emotions. Facial expressions provide a lot of information that reflects different emotional states, and they are the most significant nonverbal communication methods. Real and fake emotional discrimination enhances the user’s experience in interactive systems and provides significant insights into security and various healthcare applications. This paper introduces a new model combining two convolutional neural networks (CNNs) designed for face landmark detection and image analysis. By merging these networks, the model outperforms current state-of-the-art algorithms in detecting fake emotions via facial expression. The proposed model is trained and validated on the SASE-FE dataset. Although fake emotions have been successfully identified in most research, there are variations in methodology, dataset, and approach. The proposed model achieves a high accuracy rate of 99%. The developed model demonstrates a significant improvement in accuracy and reliability