Date of Publication :8th May 2024

Abstract:Emotion recognition is a crucial aspect of human interaction, yet individuals with alexithymia face challenges in understanding and expressing emotions. This paper proposes a Realtime Emotion Analysis system using facial recognition technology to assist individuals with alexithymia in interpreting facial expressions. Leveraging advanced face detection algorithms such as Multi-task Cascaded Convolutional Networks (MTCNN) and visual transformers for emotion recognition, the system aims to provide accurate real-time feedback on facial expressions, thereby enhancing social interactions for neurodiverse individuals. Through the integration of diverse datasets and rigorous evaluation of performance metrics, the proposed system demonstrates promising results in supporting individuals with alexithymia in navigating social interactions effectively.

Reference :