Microsoft Research cracks the curse of the group photo

Web Log: AI can be applied to people’s faces to generate an altered yet realistic image

One problem that technology has yet to solve is The Group Photo. This social phenomenon, often seen at weddings, birthday parties and conferences, is the bane of the photographer’s existence. Yes, the bride is smiling but the groom is squinting in the sunlight, his dad is sneezing and granny’s eyes are shut.

No need for endless snaps if this new technology from Microsoft Research makes its way into mainstream digital cameras and smartphones. A breakthrough in computer vision AI means that deep learning can be applied to people's faces or other categories like birds or flowers to generate an altered yet realistic image.

For example, you could take a picture of someone scowling and transform it into a completely convincing version with a beaming smile that outperforms the very best Photoshop attempts and even tricks the subject of the photo.

“What we have here is technology that can synthesize faces while preserving the identity of the faces we generate, in a controllable fashion. With our technology, the great thing is that I could literally render a smiling face for each of the participants in the shot,” explains Gang Hua, principal researcher at Microsoft Research in Redmond, Washington.