Chuck Bednar for redOrbit.com – @BednarChuck
If you’re the type of person who likes to show off the incredible meals that you’ve been cooking or eating, beware: Google is reportedly working on a new app that will reveal how many calories are in that food porn you just posted on Instagram.
According to Mashable, the app will be called Im2Calories, and Google introduced it last week at Boston’s Rework Deep Learning Summit. Using artificial intelligence, this new app will reportedly be able to determine a meal’s caloric content simply by analyzing a photo.
During the event, Google research scientist Kevin Murphy called the project a marriage of visual analysis, “in this case, determining the depth of each pixel in an image,” and pattern recognition, according to Popular Science. Despite the fact that it analyzes pixels, it also can work with lower quality images, and the algorithms are expected to become more accurate over time.
Designed for portion control, research purposes
During last week’s conference, Murphy demonstrated how the AI system could look at a photo of two eggs, two pancakes, and three strips of bacon, then calculate the approximate amount of calories in that breakfast by gauging the size of each piece of foot in relation to the plate, plus any condiments or toppings used to enhance the flavor of the meal, of course.
The goal, he told Popular Science, isn’t to fat-shame people or make them feel bad about what they’re eating. Rather, he wants to make it easier for Instagram users to maintain a food diary, using technology to automatically identify what they’re eating so that the information does not need to be manually input into an app, and to eliminate the guesswork over serving sizes.
Murphy said that the process is semi-automated, meaning that the user can correct the software by using dropdowns if the system misreads the contents of the picture. Google has filed patents related to the technology, and Murphy said that the technology should not only benefit people concerned about portion control, but also doctors and researchers who could use the data (once it reaches accuracy thresholds) in aggregate for their studies.
“To me, it’s obvious that people really want this and this is really useful,” added Murphy. “Maybe we get the calories off by 20 percent. It doesn’t matter. We’re going to average over a week or a month or a year. And now we can start to potentially join information from multiple people and start to do population level statistics we can start to potentially join information from multiple people and start to do population level statistics.”
—–
Follow redOrbit on Twitter, Facebook, Google+, Instagram and Pinterest.
Comments