Developed a desktop application that captures real-time video from a webcam and uses computer vision algorithms to detect and classify the user's facial emotion. The emotion detection pipeline processes each frame to identify expressions such as happiness, sadness, surprise, and neutral states.

Integrated a music recommendation engine that matches the detected emotion to a curated selection of Taylor Swift songs. The system maps emotion categories to song attributes like tempo, key, and lyrical sentiment to provide contextually appropriate recommendations.

Applied virtual makeup using facial landmark tracking with Dlib's 68-point face detector. AR techniques overlay cosmetic effects, such as lipstick, eyeliner, and blush, in real-time, accurately following the user's facial movements and expressions.

Built entirely in Python, leveraging OpenCV for video processing and face detection, Dlib for facial landmark extraction, and custom logic for emotion-to-song mapping and AR rendering.

Python • OpenCV • Dlib • Facial Landmark Detection • Emotion Classification • Augmented Reality