Google To Launch App That Counts Calories On Instagram Food Photos

Google is currently working on an app that will actually count the calories in pictures of food people post on Instagram.

The leading company in Artificial Intelligence is now applying the technology to a habit almost everyone is addicted to: posting pictures of food, reports Popular Science.

According to the site, Google acquired a tech company in 2014 named DeepMind, which focuses on deep learning. 

Deep learning enables a device to "learn" and improve over time by gathering and analyzing information that is already available on the net.

Deep learning has a lot of potential applications once fully researched and developed. Google is already working on using the technology in the latest app it is planning on releasing, named Im2Calories.

According to The Guardian, the new app was announced at the Rework Deep Learning Summit by Google researcher Kevin P. Murphy.

The end goal for the app is that users can use it as a food diary, since it can scan a picture of food posted on Instagram and count the dish's calories, explains Business Insider.

To count the calories the app uses algorithms to analyze the photos of food uploaded on Instagram and then estimates the number of calories found in the photos, explains Time.

The technology is not fully developed yet. Google hopes that over time the app will be able to "deep learn" so it can run a more accurate analysis of food calories.

Murphy says, "If it only works 30 percent of the time, it's enough that people will start using it, we'll collect data, and it'll get better over time."

"To me it's obvious that people really want this and this is really useful. Ok fine, maybe we get the calories off by 20%. It doesn't matter. We're going to average over a week or a month or a year," adds Murphy.

Counting calories seems to be only the beginning for the use of deep learning. According to Murphy if the Im2Calories app is fully developed, deep learning technology can be applied to other forms of information.

"Now we can start to potentially join information from multiple people and start to do population-level statistics. I have colleagues and public health, and they really want this stuff," he says.

Murphy gave a possible example of the future use of deep learning technology saying, "Suppose we did street scene analysis...We want to do things like localize cars, count the cars, get attributes of the cars..."

"Then we can do things like traffic scene analysis, predict where the most likely parking spot is. And since this is all learned from data, the technology is the same, you just change the data."