In less than a minute, software can predict what a child will look like when he/she grows up.
Developed by the University of Washington and funded by Google and Intel, the software automatically generates images of a young child’s face (photos that begin at age 5 work best) as it ages to age 80. This is the first fully automated approach for aging babies to adults that works with variable lighting, expressions and poses.
The software corrects for tilted faces, turned heads and inconsistent lighting, then averages thousands of random Internet faces of the same age and gender, and they are applied to the child's face to project his/her againg face.
The renderings are so good that they are hard to discern from real photos.
The researchers hope the software can factor in ethnicity, hair whitening and wrinkles in the future.