You will have seen photos on Facebook, Google Plus or other social networks that testify to the location where they were taken. The geotags use the positioning function. But the future is coming…
Today, MIT Tech Review announced details of a new project led by Tobias Weyand, an artificial intelligence expert intelligenceof Google. The project "breaks” φωτογραφίες μέχρι σε επίπεδο pixel και στη συνέχεια προσπαθεί να παραβάλει αυτά τα pixels με τεράστια βιβλιοcase images to determine where they fit best.
It is the digital equivalent of the puzzle where you have a piece and try to find its exact position. Only in this case, there are hundreds of millions of alternatives.
The sophisticated AI, called PlaNet, uses over 90 millions of images with geographic coordinates all over the globe. In a test with 2,3 million images, PlaNet was able to determine the country of origin 28,4% and time and continent of origin in 48% of cases.
Although these Results they are not perfect, the first numbers are significantly better than the results of human trials.