Microsoft’s Project Adam takes image recognition to the next level

You may have used Google’s reverse image search’s recognition software to call shenanigans on a picture someone claimed to have taken themselves, or stalk someone on FB.

But Microsoft’s Project Adam looks to expand the use of image recognition, and has developed its software to some impressive lengths.

First off, Adam is allegedly 50 times faster and up to two times as accurate as previous image recognition software. For example, let’s say you’re trying to use image recognition on a car, Project Adam will be able to tell you the brand and make and model year of the vehicle. Pretty nifty. The folks behind Adam’s software are also looking to add text and speech recognition to help make Project Adam learn and adapt based on the different environments it is exposed to:

“It automatically learns how to extract features from these images, so that when you show it an image that it has never seen before, it can accurately categorize it in one of the categories that you’ve already taught it,” says Partner Research Manager Trishul Chilimbi

Adam is also capable of taking a look at the food you’re eating and being able to give you the nutritional content, more or less, of the grub.

What makes Adam unique is that the software doesn’t look at the picture as a whole, but examines sections of the image. This affords Adam the ability to be more accurate than other image recognition technologies.

There hasn’t been any word yet as to when we can expect Adam in our devices, but it definitely sounds like a cool concept if it works as advertised.

Source | Via