Amazon recently released its first smartphone, the Fire Phone, and one of the device's most talked about features was Firefly which uses the phone's camera to recognize objects and then finds them in a database. For most physical products you'll then get the immediate option to purchase them on Amazon.

Image recognition makes obvious sense for online retailers, though now Microsoft has announced Project Adam which aims to take image recognition to new levels. Project Adam uses a "Deep Learning" system that works similarly to the human brain and, so Microsoft claims, makes Project Adam world class in performance, scaling and accuracy for visual recognition tasks and thus the world's best photograph classifier.

A gigantic data set and over 2 billion "neuronal" connections between data points should double the accuracy of current state-of-the-art systems and make Microsoft's system up to 50 times faster. 

"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.

The system's neuronal connections are used to analyze small areas of an image rather than the entire frame. By doing this the system can be more precise at the recognition of facial features or textures and therefore improve image classification.

No indication is given as to when we can expect the technology to be available in a device but in the video below Microsoft describes a couple of potential use cases. For example Project Adam will allow you to take a picture of your food and then give you the nutritional information, just to confirm that donuts are bad for you. Maybe more usefully the system will also identify certain diseases by analyzing pictures of the visible symptoms on your body.  Watch the video below to learn more.

Via: GSM Arena