What does the Earth look like to Facebook's Image Recognition Engine?

What if you applied the same technology used to identify and automatically tag you and your friends in your Facebook photos to understand the changing world we live in?

Today Facebook publicly discussed doing exactly that using DigitalGlobe's Geospatial Big Data initiative along with Facebook’s image-recognition engine and a tailored convolutional neural network in order to extract image features.

The world is changing, but what is even more exciting is that our ways to "see" the world are changing as well.

I truly believe this is only just the beginning of a real revolution where we find ways to not only see a better world, but contribute to making a better world as well.

A great overview by Dr. Shay Har-Noy on the DigitalGlobe blog provides a more in-depth overview of DigitalGlobe's support to this effort...

The post by the Facebook Connectivity Lab staff that led the effort is equally enlightening and detailed...

What gets me up every morning (and keeps me sitting up at night typing up posts like this!) is that DigitalGlobe is opening up the opportunity for anyone to unlock these types of insights from our (and other content providers) data via our GBDX Platform.

By making the worlds largest data set of satellite imagery accessible and building an ecosystem of algorithm, machine learning, and computer vision capabilities on top of it, it won't only be Facebook who will be able to obtain these insights.

So, let me ask you...

What questions would you answer if you could?