Declassifier processes pictures using the YOLO computer vision algorithm. Instead of showing the program's prediction, the picture is overlayed with images from COCO, the training dataset from which the algorithm learned in the first place.
The data by which machine learning algorithms learn to make predictions is hardly ever shown, let alone credited. By doing both, Declassifier exposes the myth of magically intelligent machines, instead applauding the photographers who made the technical achievement possible. In fact, when showing the actual training pictures, credit is not only due but mandatory.