U.S. researchers developed a software program, which is claimed to be able to automatically generate sequential images of the future, predicting how the face of a child will change from decade to decade, as he will be growing from three years old to… 83.
The research, funded by the Google company and the biggest processor manufacturer Intel, was conducted by scientists at the University of Washington, led by computer science professor Steven Seitz.
The software can be run on a single PC and takes just a minute to produce a series of portraits of the future look of a three-year child. The algorithms are based on a comparative analysis of thousands of photos that show the same person at different periods of his life, as he is aging, thus allowing the program to make predictions about how the person can change over time.
The tests so far show that the program is quite successful and produces convincing images to the point that people find it difficult to distinguish whether a photo was taken in reality or created by the computer based on a child photo of the same person. The public will soon have the opportunity to test the program with the help of a web camera, say the researchers.
Copyright © 2012-2020 Learning Mind. All rights reserved. For permission to reprint, contact us.