X

No more selfie distortion? Tool helps you put your real face forward

Bad case of selfie face? A photo-editing tool out of Princeton automatically adjusts camera distance to remove the fallout of getting in too close.

Leslie Katz Former Culture Editor
Leslie Katz led a team that explored the intersection of tech and culture, plus all manner of awe-inspiring science, from space to AI and archaeology. When she's not smithing words, she's probably playing online word games, tending to her garden or referring to herself in the third person.
Credentials
  • Third place film critic, 2021 LA Press Club National Arts & Entertainment Journalism Awards
Leslie Katz
2 min read

Those of us who don't have pro photographers following us around snapping portraits from the most flattering angles often rely on shots taken via awkwardly placed selfie arm. But camera proximity can make your nose look bigger, your ears look smaller and your forehead look more slope-ey.

Didn't realize selfies could do all that? Happy Monday!

Enlarge Image
Princeton School of Engineering and Applied Science

The good news: computer scientists from Princeton University have developed a tool that corrects that common brand of selfie photo distortion that comes from getting up so close in your own mug.

The tool, presented at the Siggraph 2016 computer graphics conference last week, can modify faces to look like they were photographed from farther away. It can also alter the subject's apparent pose, making it appear as if the camera was placed higher, lower or at an angle.

"Although it is the age of the selfie, many people are unaware of how much these self-portraits do not really look like the person being photographed because the camera is way too close," Ohad Fried, a Ph.D. candidate in Princeton's Department of Computer Science and lead developer of the new method, said in a statement. "Now that people can edit so many aspects of a photo right on their phones, we wanted to provide a quick way to edit faces that maintains realism."

The Princeton team started with a database of digital 3D heads compiled by Zhejiang University in China. They then turned to a program made available by Carnegie Mellon University that identifies nearly six dozen reference points across the face, such as the corners of the eyes and top of the head and chin, in a selfie. From there, they figured out a way to adjust the 3D head model so it corresponded to the points detected on the face in a way that looks natural instead of distorted.

The editing methods "do not create new people," reads the full paper (PDF). "Rather they show the same people under different viewing conditions. In that sense, they are the post-processing equivalent of a portrait photographer making a different decision about the composition."

The paper offers a much more detailed explanation of the tool, including phrases like "the standard tensor-vector multiplication in the i-th dimension." But if you just want to know when you'll be able to press a button to make your forehead look less slope-ey, the Princeton team says it has some refining to do before the tool becomes commercially available.

"We are happy with what we achieved so far, but we look forward to learning how we can make these selfie transformations appear even more realistic," Fried said. Now for a cure to selfie elbow...