Emotionscan: Facial recognization and mood detection system

Degree

Master of Science in Computer Science

Department

Department of Computer Science

School

School of Mathematics and Computer Science (SMCS)

Date of Submission

Fall 2024

Supervisor

Dr. Muhammad Sarim, Visiting Faculty, Department of Computer Science, School of Mathematics and Computer Science (SMCS)

Keywords

Emotion recognition, CNN, DDAMFN++, AffectNet, deep learning, Spotify integration, Emotional well-being

Abstract

Human emotions, as reflected through facial expressions, provide a vital medium for understanding psychological states and overall well-being. This research focuses on developing a real-time mood detection system utilizing advanced deep learning techniques, including Convolutional Neural Networks (CNNs) and the Dynamic Dual Attention Mixed Feature Network++ (DDAMFN++) model. By leveraging the AffectNet dataset, a comprehensive collection of annotated facial expressions, the system identifies emotions such as happiness, sadness, anger, and neutrality with remarkable accuracy. The DDAMFN++ model incorporates dynamic attention mechanisms to effectively analyze spatial and temporal features, while CNNs enable precise feature extraction from facial expressions. Integrated with OpenCV for real-time video processing and Spotify's API for mood-based music recommendations, the system bridges cutting-edge technology with practical applications. A robust token management framework supports seamless integration with external services. This project represents a significant step toward creating responsive digital environments that enhance emotional well-being. By delivering accurate emotion detection and actionable interventions, it provides a novel solution for real-time mental health support and personalized user experience

Document Type

Restricted Access

Submission Type

Research Project

The full text of this document is only accessible to authorized users.

Share

COinS