Jun 3, 2015 · 2 minutes

Google has finally found a use for the disgusting amount of food porn shared on Instagram, Facebook, and other social networks after every single meal: visual analysis that's supposed to figure out how many calories were in a dish.

The project has been given the distinctly unappetizing "Im2Calories" moniker. It's said to rely on technology from the DeepMind artificial intelligence startup acquired by Google in 2014, and right now it's little more than an experiment -- albeit a fascinating one.

That's because the deep-learning technologies powering this project will eventually be used for something besides counting calories. As Popular Science, which first revealed Im2Calories, explains in its report on Google's efforts:

Obesity is a scourge, and deserves all the sophisticated deployment of semantic image segmentation and deep neural networks that Google can muster. But robot cars that instinctively know which block is most likely to have a free parking spot, ten minutes from now? It's not surprising that deep learning is drawing so much interest from Silicon Valley. If anything, it's a surprise that it's taken this long.
Yet that doesn't mean Im2Calories isn't worth considering on its own merits. Like that report points out, obesity is a serious problem in the United States, and it's hard not to wonder how a project like this might be able to help people.

Other services focused on healthy eating have one big drawback: people aren't using them. Entering a food into a service offered by FitBit or MyFitnessPal requires users to create a new habit to go with every meal. That's not true of Im2Calories.

Instead, it takes a behavior most people already embrace -- sharing images of whatever happens to be on their plates -- and makes it useful for more than just bragging about having a delicious beef bourguignon at a fancy restaurant.

Which isn't to say that Im2Calories isn't flawed, both as a concept and in its current form. Google admits that its calorie counts can be off by as much as 20 percent, and the Verge notes that counting calories isn't always helpful anyway:

There's no questioning Google's ability to collect and process data from a large number of users, but the company would still have to solve one big problem for something like Im2Calories to work: calorie counting is unreliable. A major study published in The New England Journal of Medicine in 2011 found that quality of food is a more important contributing factor than quantity. 'Conventional wisdom — to eat everything in moderation, eat fewer calories, and avoid fatty foods — isn’t the best approach,' Dr. Dariush Mozaffarian, the lead author of the study, told The New York Times. 'Just counting calories won’t matter much unless you look at the kinds of calories you’re eating.'
That might be true. But encouraging some level of awareness for what people are putting in their mouths, whether it's through a calorie counter or a visual representation of what they've been eating, could still lead to some changes.

So congratulations, foodies! Your culinary exhibitionism might help people.

[illustration by Brad Jonas]