Researchers at the University of Washington have created a way to take a collection of photos of a person and automatically turn them into an animation.
It can make a face appear to age over time, for example, or make the person’s expression gradually change from a smile to a frown.
“I have 10,000 photos of my five-year-old son, taken over every possible expression,” says Steve Seitz, a UW professor of computer science and engineering and engineer in Google’s Seattle office.
“I would like to visualize how he changes over time, be able to see all the expressions he makes, be able to see him in 3-D or animate him from the photos.”
The software can work with tagged photos from the web or personal collections. It locates the face and major features, then aligns them, choosing photos with similar expressions so the transitions are smooth.
One example video uses photos of a Google employee’s daughter taken from birth to age twenty. The owner scanned the older photos to create digital versions, tagged them with the subject’s name and manually added the dates. The result is a movie in which the subject ages two decades in less than a minute.
The researchers say that for modern babies, who are digitally chronicled from before birth (now, there’s a thought), such films will be much easier to create.
“There’s been a lot of interest in the computer vision community in modeling faces, but almost all of the projects focus on specially acquired photos, taken under carefully controlled conditions,” says Seitz.
“This is one of the first papers to focus on unstructured photo collections, taken under different conditions, of the type that you would find in iPhoto or Facebook.”