A powerful real-time face detection desktop application built using Qt (C++), OpenCV DNN, and modern UI principles. It supports webcam, local video files, manual face capture, and thumbnail previews, making it ideal for educational projects, research, or internship showcases.
- 🎥 Live webcam streaming with real-time face detection
- 📁 Load and analyze local video files
- 🤖 DNN-based face detection using OpenCV’s
res10_300x300_ssdmodel - 🖼️ Thumbnail previews of manually captured faces
- 💾 Saves cropped face images to
faces/folder with timestamped filenames - 🎛️ Responsive modern GUI built with Qt Designer
- 🟢 Start / 🔴 Stop / 📸 Capture buttons
- ✨ Status label + face count label
- ✅ Fully working on macOS (Apple Silicon / Intel)
| Live Detection | Face Capture | Thumbnails |
|---|---|---|
 |
 |
 |
- Language: C++
- GUI: Qt 6 / Qt Widgets
- Face Detection: OpenCV 4.x + DNN module
- Build System: CMake
- Platform: macOS (tested on Qt 6.9.1)
FaceDetector/ ├── main.cpp ├── mainwindow.cpp / .h / .ui ├── haarcascade_frontalface_default.xml (optional) ├── deploy.prototxt ├── res10_300x300_ssd_iter_140000.caffemodel ├── faces/ # Cropped face images saved here ├── CMakeLists.txt └── README.md
git clone https://github.com/manavmax/FaceDetector.git
cd FaceDetectorFile → Open Project → CMakeLists.txt Make sure OpenCV is linked in Kit / CMake configuration
Click the green
Ensure the following two files are placed in the same folder as the executable:
deploy.prototxt res10_300x300_ssd_iter_140000.caffemodel
If running from Qt Creator, copy them to the build/ directory:</small>
cp deploy.prototxt res10_300x300_ssd_iter_140000.caffemodel ./build/</small>
Manav Sharma
✅ Project is complete and fully functional.