Selfies with science? Researchers at the University of Waterloo have created an app to help you take better selfies.
One of the scourges of the social media landscape, the poorly shot selfie has for too long been clogging up our Snapchat and Instagram accounts, momentarily engaging us with their spontaneity and joie de vivre (Hey! Look! I’m at Buckingham Palace!) but then quickly turning us off by being so badly lit and ill-framed (Dude, there’s a flagpole sticking out of your head, in front of Buckingham Palace).
Fear not, as now there’s an app to help you along the path to a PhD in cameraphone self-portraiture, one that teaches you on the fly how to get that better shot with which to thrill all of your internet fans — something that researchers say has become a more common method of communication.
“Selfies have increasingly become a normal way for people to express themselves and their experiences, only not all selfies are created equal,” said Dan Vogel, a professor of computer science at the University of Waterloo in Waterloo, Ontario, in a press release. “Unlike other apps that enhance a photo after you take it, this system gives direction, meaning the user is actually learning why their photo will be better.”
The new algorithm from Vogel and co-creator Qifan Li, a former Master’s student at Waterloo, focuses on three compositional features of a great selfie: lighting, face position and face size. The researchers used crowdsourcing to pick through hundreds of computer generated photos of six “average-looking” people taken from different angles and with different lighting. The project produced 2,700 crowdsourced aesthetic ratings, which were then crunched into the algorithm that guides the user towards better positioning and lighting through a live preview on their hand-held device.
To test their product, the researchers had real people take photos both with the app and without and had the results again rated online. “Our results show that our system improved selfie photograph aesthetics by 26 per cent,” say the study’s authors, whose work was recently presented at the 2017 ACM Conference on Designing Interactive Systems in Edinburgh, Scotland.
“This is just the beginning of what is possible,” said 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.”
On that note, the sky seems to be where the selfie market is headed, as reports are that Snapchat is looking to buy Chinese drone company, Zero Zero Robotics, makers of the Hover Camera Passport. Allowing the average Joe or Jane to become even more the centre of their own attention, the Hover Camera has body tracking, face tracking and orbiting capabilities, all in a tiny package (it’s paperback novel-sized when folded up in its case and weighs only 242 grams).
Snapchat, which went public in March, is reportedly hoping to distinguish itself from other imaging platforms by jumping into the hardware market. The deal for Zero Zero Robotics is said to be in the $150 to $200 million range.