Since 1984, forensic artists at the National Center for Missing and Exploited Children have created almost 6,000 age-progression images of missing children.
Over 1,300 of those missing children have been found with the help of these “aged” images.
The center currently employs four forensic artists who work to update the photos of missing children every two years.
Missing children under the age of 18 get their photo updated every two years. However, once the child turns 18, their photo is only updated every five years.
The age-progression images help show the general public what the missing child would likely look like today.
The work of forensic artists has become so good that it has helped solved old missing persons cases from decades ago.
However, their work can be time consuming and subjective since these images are created manually and rely on old pictures of the child as well as pictures of family members and photo editing software.
Just this year a Jane Doe case from the 1980s was solved due to the work of digital forensic artists at the Federal Bureau of Investigation.
FBI artists created a three dimensional image from the remains of a woman’s skull. The woman went missing in the 80s and was never identified.
The forensic image was made public and a family member recognized the woman. This breakthrough helped close this missing persons case and brought closure for the family who lost her.
At the moment, the creation of age-progression composites can take some time since there can be a waiting list for the work of good forensic artists.
In addition, the work of forensic artists is subjective since it is based on what the artist thinks the person would look like at their current age.
However, age-progression software that can produce these forensic images automatically is getting better, cheaper and faster.
Also, photos that are aged by a computer program are less subjective because the software can factor in facial features from a multitude of other pictures that depict different races and genders at different ages.
The picture below shows actual photos (right) next to computer-aged photos (left) for a person from age 3 to age 16.
Last year, researchers at the University of Washington announced the development of image software that can process a child’s picture and predict what the child will likely look like at any age all the way up to 80 years old.
The software uses an aging algorithm that scans thousands of random online images of people from different ages and genders and “learns” how to age a person based on the average changes from the scanned photos.
The program can then take what it has learned from previous images and apply it to any photo of a baby or child.
The researchers then compared their computer-aged photos with dozens of known people who had been photographed over the course of a lifetime.
The software’s age-progression images are so lifelike that people cannot tell which photos are real and which photos have been “aged” by the software.
The program can run on an average desktop computer and takes only 30 seconds to produce “aged” images for one person’s face.
The researchers believe that this software could aid law enforcement in solving the large number of missing persons cases around the world.
Although this program is not available to the public yet, the researchers plan to offer a smartphone app and an online tool where people can upload and process their own photos.
You can find out more about this age-progression software in the video below or by visiting the University of Washington site.