Are your selfies the best that they can be? Unless you've done a lot of research into what makes a great selfie, they probably aren't. Fortunately enough, however, computer scientists at Canada's University of Waterloo have done that research – and they've used it to create an app that guides users in taking the best selfies possible.
Prof. Dan Vogel and former Master's student Qifan Li started with 3D digital scans of a variety of average-looking people. They then used a computer to take "virtual selfies" of those faces, in which they controlled a virtual smartphone camera. Variables that they could change included lighting direction, face position within the shot, and face size.
The researchers then utilized the Mechanical Turk online crowdsourcing service, getting thousands of users to vote on which virtual selfies they liked best. Based on qualities that the winning images had in common with one another, an algorithm was created to guide people through the process of taking real-life selfies.
Vogel and Li next had people take selfies using a phone's standard camera app, and then again as guided by an app based on the algorithm. When the resulting shots were voted on online, the app-guided photos were rated 26 percent higher.
"This is just the beginning of what is possible," says Vogel. "We can expand the variables to include variables aspects such as hairstyle, types of smile or even the outfit you wear. When it comes to teaching people to take better selfies, the sky's the limit."
There's more information in the following video.
Source: University of Waterloo
Want a cleaner, faster loading and ad free reading experience?
Try New Atlas Plus. Learn more