Google Cloud Vision API

Google Opens Image Recognition API To Developers

December 3, 2015 - Written By John Anon

Google Photos is technically one of the newer apps available from Google. At least, it is one of the newer big apps from the company and it is one which instantly landed with impact. In fact, since its arrival, Google Photos has gained a great deal of popularity and usage. One of the reasons for why the app has become so popular is due to the additional features on offer with the app and especially compared to other photo apps. One of those features in particular, is the ability for Google Photos to identify certain aspects like animals, places and even people. If you ever wondered how the app achieves this aspect recognition, then part of the intelligence comes from Google’s Cloud Vision API.

This API is part of the underlying software which is able to capture and identify a user, a dog, or otherwise and can translate that information into something more meaningful, like recognition. Instead of just identifying an image, the technology can identify the contents of an image. Following on from this and yesterday Google announced that the Cloud Vision API was effectively being opened to the developer community. As such, developers can now use the Cloud Vision API in their own applications to achieve much of the same results.

If you are wondering what could be achieved with the Cloud Vision API (beyond a photo app like Google Photos), well, Google also offered an insight into that too. The video below shows a robot which is programmed with the same information and how the robot can effectively gauge a situation or a person and make a decision. In the video, this is explained as the robot knows to approach when someone is smiling and to stay away when they are surprised or angry. Likewise, the robot is also able to understand prompts when shown. Like for instance, a banana. So, it does seem that the Cloud Vision API does look to be something which has a far greater potential than simply a means to catalog your images in a meaningful way and now that it has been released for third party use, it is likely will provide the backbone for some interesting products and features going forward.