Basic webscraping with Selenium and Beautiful Soup to be used for building ML training face image datasets. The double p is intentional.

Depending on your setup, you might be missing some Python modules, just

sudo pip install -r requirements.txt

You might also need to change the browser driver executable, included in the repo are web drivers for chrome on Windows and Linux. Find yours at (https://chromedriver.storage.googleapis.com/index.html), download and unzip, then add to the project folder or add to system path. If you are using Firefox, the process is similar, just google for Selenium python Firefox.

Example usage:

python scrapper.py -p google -m "brad pitt"

The app has both a CLI parser, as well as a graphical user interface. Check out --help for details.

There are 3 modes of use: automated search (default), manual search (-m "search term") and filtering/cropping (-f on) of faces. Extra functionality built to enhance the results of these modes is available (e.g. automatic name list generation or fine parameter settings).

The app searches for, downloades and processes Google Image Search results and IMDb pictures of actors available freely online and is intented for academic use only.