Google is always trying to advance technology, and in the future the Silicon Valley company wants diners to count calories simply by taking photos of their meals. Google's artificial intelligence team is working on a project called Im2Calories, which uses "deep learning" to determine the caloric value of food items, reports Popular Science. The end goal is to make it easier to keep food diaries and help fight obesity, which is a major problem in America.
Because the technology learns with experience, Im2Calories should continue to get more accurate at identifying specific items and how many calories they contain. It analyzes pixels in photos to determine what, exactly, it's looking at, and then pairs that information with data on the typical number of calories associated with specific items. The photos don't need to be high quality to get results.
"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," Google research scientist Kevin Murphy said at a recent presentation during the Rework Deep Learning Summit.
It's not Google's first foray into the world of dining. The company has been making it easier for consumers to find food and drinks for a while, integrating Open Table into its maps application, offering instructions on making cocktails, and making it possible to order delivery directly from its search engine.