Microsoft's new app uses machine learning to recognise dog breeds

The use of machine learning means Fetch! will get better based on how much people use it.

Published Date
12 - Feb - 2016
| Last Updated
12 - Feb - 2016
 
Microsoft's new app uses machine learning to recognise dog breeds

Microsoft’s newest app for iPhones, called Fetch, can guess a dog’s breed. The app uses machine learning to recognise dog breeds. It asks the user to take a picture of a dog and utilises facial recognition software to recognise its breed. The app is available for iPhone and on what-dog.net. Sounds like a weird idea for a software giant?  Well it’s not. As Mitch Goldberg, development director at Microsoft Research, explains, “There was an interest in creating a framework that would allow you to take a domain - in our case, dogs - and recognise numerous classes, such as breeds. We were interested in enabling an app to allow you to make object recognition extraordinary, fun and surprising.”

It’s evident that Microsoft is basically putting machine learning to good use here. The app may not seem like much, but it’s yet another step that Microsoft has taken towards better use of machine learning. To understand why, one needs to realise that machine learning algorithms get better as you throw more data at them, which is exactly what apps like Fetch!, Microsoft’s Mimick Alarm and even Cortana for Android are meant for. On the face of it, these seem like simple apps, but they get Microsoft a lot of data, which in the long term, will help the company grow its products.

“We wanted to bring artificial intelligence to the canine world. We wanted to show that object recognition is something anyone could understand and interact with,” added Goldberg. According to the company’s blog post, Fetch! is designed for repeated use and will get better based on how much it is use. It is worth noting that Microsoft isn’t just talking about your usage, but the total usage of the app all over the world.

Digit NewsDeskDigit NewsDesk

The guy who answered the question 'What are you doing?' with 'Nothing'.