This repository contains code to use the google cloud face API. There is an associated post, available here:

https://folk.uio.no/jeanra/Informatics/GoogleCloudExampleFaceAPI.html

You will need to set up a Google Cloud account, create a project, enable the API, and export credentials as indicated in the post to be able to use the code.

This is developed for Ubuntu 16.04 with Python 2.7. My computer has a webcam available on /dev/video0. You can adapt the code to use another video input.

To get the code ready to work:

• Copy the code on your computer, for example by cloning this repo.
• Make sure to have the necessary packages: ffmpeg, v4l-utils.
• Make sure you have the necessary python modules: scipy, matplotlib, google, a (correctly) installed PIL.
• Create and set up your google cloud account, project and credentials: see the associated post.
• Adapt the detect_faces.sh script with the right path to your JSON credential.
• Make the script executable: chmod +x detect_faces.sh

Simply:

./detect_faces.sh

Have fun!