Facelab is a graphical front-end for doing image recognition and processing, primarily on faces. It is written in Python and OpenCV, and is designed to be extensible.
Each graphical element is a 'Widget' object, that runs in a separate thread. Widgets can be added in and removed with minimal changes to the code - they are all loaded at runtime. All the graphical operations are handled by OpenCV for speed.
Currently, the code comprises:
- Histogram plot of pixel intensities
- Frame rate (FPS) and history plot
- Elapsed time
- Haar cascade face detector
- Picture-in-picture view of detector output
Run the code by executing main.py. Requires Python 3, numpy and OpenCV 3+.
Contributions welcome.