The New York Time is working with Google to digitize its huge archive of millions of images.
As the Google team states, The New York Time stores between 5 and 7 million photographs, with great historical value, since the collection dates back to the late 19th century. To preserve the integrity of this file, not only is it necessary to keep the images, but all the data they contain, because this information allows us to have the exact historical context of over 100 years of world events.
Thanks to the potential of Google's AI, they will be able to digitize each of the details, both images and handwritten notes, and then, classify the semantic information, to generate the corresponding context. Google illustrates this point with the following image:
The first image would lose its historical context if the user does not take into account all the information that is added to the back, such as the clipping of titles on the event of the photograph.
In addition, Google will also implement its object recognition tool, which analyzes every detail of the image. For example, in the image above, you can identify the Pennsylvania station, even if the photograph was taken in 1942. Combining all this information, artificial intelligence can convert a simple photograph into a wealth of information, inserting the date, the position, among other details.
Although this large file will not be available to the public, but will be for internal consultation, it will allow us to appreciate the great potential of Google's IA.