- Run pip install opencv-python or change directory to C:\Python27\Scripts\ and run pip.exe install opencv-python
- Download lfw from http://vis-www.cs.umass.edu/lfw/lfw.tgz (or any set of images you want to use)
- Create "recognizedFaces" folder inside the folder you unziped this git
- Run python server.py to start the server
- Run python client.py 8 "path/to/the/image_set" to start the transfer of images from client to server
- Wait till the transfer is done and the face recognition will automatically start
- Wait till the processing is done and the recognized faces images will appear inside recognizedFaces folder
Obs: The second parameter in step 5 is the number of processors you want to use, in my case was 8.