Using advanced image-recognition technology to work out what food you're eating, Im2Calories estimates what the calorie count adds up to.
The app, unveiled at this week at a tech conference in Boston, uses algorithms to analyze photos to calculate each total. The underlying technology comes from DeepMind, a company Google acquired last year for $400 million.
The size of each piece of food is worked out by comparing it to the rest of the plate - with condiments in the picture even being added by the app to the calorie calculations.
Popular Science have reported that the app will learn over time, with users being asked to input food details for any items that the app cannot recognise.
Google research scientist Kevin Murphy explained that in the beginning the calorie count could out by anything up to 20 per cent but added that the technology will become more accurate each time it is used.
He said: "If it only works 30% 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. 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."
Given the amount of people who regularly add pictures of their meals to social media sites such as Facebook and Instagram, the technology could become incredibly popular.
However, there are some drawbacks. The app cannot differentiate between the differing attributes of each ingredient used, so if a low-fat alternative ingredient is used the app will likely overestimate the calories.
There are also concerns about calorie counting itself as a viable science.
The app is still in development and has not yet been released, though Google are reported to have already filed a patent application.
