We’ve all seen them – the food pics are everywhere. Instagram has basically become a food porn haven, with everyone sharing their delicious lunch or snack. But those pics could actually yield valuable information, and tell you how many calories you’re eating.
I honestly don’t think we need to give people more motivation to post food pictures (Instagram could serious be so much better), but it just seems like there’s no good way to stop people, so if you can’t stop them… help them. That seems to be the thought process behind one of Google’s recently announced projects. According to CNET, Kevin P. Murphy, a researcher who works with Google’s research lab and specializes in artificial intelligence, algorithms and theory, said the new AI uses “the depth of each pixel in an image” and “sophisticated deep-learning algorithms” to identify food, judge its size and come up with a calorie count. The photos don’t even need to be high-definition for the thing to work. The AI will not get it right at first, but the algorithm will be built in such a way that it will learn and adapt, and it will eventually start getting it right.
The point isn’t to embarrass people or brag about your healthy diet – it’s to enable us to keep a sort of food diary where we easily keep track of our calorie intake. Sure, there are some good calorie calculators out there, but it’s all portion-based; in other words, it can be very difficult to know how big your portion is. However, there is more than just weight watching at stake. Murphy noted at the presentation that his big hope is that the technology could be some day used to monitor things like traffic and spot ways to improve it (for example helping drivers find a parking spot quicker). But in the mean time, obesity remains a big problem (especially in the US), and I feel confident that a commercial version of Im2Calories would be widely successful.
“To me it’s obvious that people really want this and this is really useful,” he said. “Ok fine, 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. I have colleagues in epidemiology and public health, and they really want this stuff.”
However, it may still be a few years before this technology actually gets implemented. While the algorithms are constantly improved, Google spokesman Jason Freidenfelds said that they are still at a research phase and no actual products are planned for the near future.
Enjoyed this article? Join 40,000+ subscribers to the ZME Science newsletter. Subscribe now!