Google buys image recognition startup Moodstocks

Moodstocks developed machine learning technology to recognize images and objects on mobile devices

Machine learning is creating some pretty cool technology, including speech recognitiion, improved health benefits and the self-driving car. It's also allowed for image and object recognition, where a computer will be able to scan a photo, or even objects in the real world, and immediately identify them. 

There's all sorts of uses for this. For Google, it means the ability to categorize data, leading to smarter, better search results. That's why the company has acquired Moodstocks, a  provider of an API and cross-platform SDK, which allows developers to integrate scanning into their apps. 

Founded in 2008, the Paris-based Moodstocks initially focused on on-device image recognition, the company has been working on extending its reach to object recognition for the past two and a half years.

No financial terms of the deal were disclosed, but in a post on its website, Moodstocks revealed, that it will be shutting down its services. 

"Today, we’re thrilled to announce that we’ve reached an agreement to join forces with Google in order to deploy our work at scale. We expect the acquisition to be completed in the next few weeks," it wrote. "Our focus will be to build great image recognition tools within Google, but rest assured that current paying Moodstocks customers will be able to use it until the end of their subscription."

The team will be coming to work at Google's R&D Center in Paris, which contributes to the development of YouTube and Chrome.

"Many Google services use machine learning to make them simpler and more useful in everyday life - such as Google Translate, Smart Reply Inbox, or the Google app," Google wrote in a blog post. 

"We have made ​​great strides in terms of visual recognition: now you can search in Google Pictures such as 'party' or 'beach' and the application will offer you good pictures without you ever needing to categorize them manually. But there is still much to do in this area."

It is unclear if Google has any plans to incorporate any of Moodstocks' technology, or how it plans to do so. VatorNews reached out to Google for more information and we will update this story if we learn more. 

This is only Google's fourth acquisition this year, after it previously bought Singapore-based communications service Pie, virtual guide Synergyse, and high-speed Internet provider Webpass. 

The company has slowed down considerably in the number of companies it has been buying, after years of being an acquisition fiend. Last year, for example, the company made 16 purchases, down from a pretty extraordinary 35 acquisitions in 2014. Not only that, but it's the only company with more than 100 in the last five years.

Object recognition

Google is not the only company with plans for object recognition. Last month Twitter bought Magic Pony, a company that uses machine learning to enhance visual processing. Twitter plans to use the technolgy to enhance its strength in live and video.

Pinterest has also taken a keen interest in this type of technology, using to help strengthen the purchasing power of its users.

In 2014 it bought VisualGraph, followed by visual organization tool Icebergs, and, in November of last year, the company debuted the ability to zoom in on and select objects in a photo, which they could then search for.

For example, if there was a picture of a table, they could zoom in on a lamp in the corner instead, to find out where to buy it.

Pinterest recently upgraded those capabilities, so that objects would be scanned automatically, but even cooler is that the company is developing technology so users can do the same thing in the real world. All users would have to do is take a picture of something, and Pinterest will give them a recommendation on where to find it. They would even do the same thing for multiple items in the same photo.

(Image source: moodstocks.com)