Description

Desktop face detection app with glowing pixel UI and OpenCV DNN backend.

How I Made It

• Built with PyQt5 to create a custom dark GUI with glow effects and responsive layout.
• Implemented OpenCV's DNN Face Detection using SSD + ResNet (Caffe model) for modern facial recognition.
• Used QGraphicsDropShadowEffect for pixel glow aesthetics on UI labels.
• Auto-saves each snapshot with incrementing names for both raw and processed frames.
• Packaged into a double-clickable app using a simple .bat launcher and Python 3.13 compatibility.

Challenges Faced

• MediaPipe was incompatible with Python 3.13 — switched to OpenCV DNN for better compatibility.
• Creating glowing text in PyQt required graphic effects, as CSS `text-shadow` isn't supported in Qt style sheets.
• Ensured the app worked across resolutions and handled webcam errors gracefully.

Key Wins

• Built a desktop AI utility with a rich UI and modern detection backend.
• Delivered responsive snapshots with clean auto-naming and a smooth user experience.
• Achieved high detection accuracy using a ResNet-SSD pipeline with OpenCV.
• Bridged creative design with machine learning for a polished user tool.