Skip to Main Content
PCMag editors select and review products independently. If you buy through affiliate links, we may earn commissions, which help support our testing.

Google RAISR Intelligently Makes Low-Res Images High Quality

By using machine learning images can be up to 75 percent smaller without losing their detail.

January 13, 2017
Google RAISR Eyes

With unlimited data plans becoming increasingly expensive, or subscribers being forced to ditch their unlimited data due to overuse, anything that can reduce the amount of data we download is welcome. This is especially true for media including images or video, and Google just delivered a major gain when it comes to viewing images online.

The clever scientists at Google Research have come up with a new technique for keeping image size to an absolute minimum without sacrificing quality. So good is this new technique that it promises to reduce the size of an image on disk by as much as 75 percent.

Google RAISR Surfer

The new technique is called RAISR, which stands for "Rapid and Accurate Image Super-Resolution." Typically, reducing the size of an image means lowering its quality or resolution. RAISR works by taking a low-resolution image and upsampling it, which basically means enhancing the detail using filtering. Anyone who's ever tried to do this manually knows that the end result looks a little blurred. RAISR avoids that thanks to machine learning.

The eyes image above is a great example, as is the horse head below. In both cases, the lower resolution is the data being worked with, and the higher resolution version being what RAISR produces after filtering. The quality gain is clear to see:

Google RAISR Horse Head

By showing RAISR a low quality image, it can intelligently upscale it to look like a high quality equivalent. Just as importantly, it does this on your device after the data-saving low resolution image has been downloaded, with the conversion happening in real-time even on mobile devices.

RAISR has been trained using low and high quality versions of images. Machine learning allows the system to figure out the best filters to recreate the high quality image using only the low quality version. What you end up with after lots of training is a system that can do the same high quality upsampling on most images without needing the high quality version for reference.

Anyone visiting Google+ since November has probably already seen images that have been tweaked by RAISR. It's currently handling over a billion images a week and saving an enormous amount of bandwidth for end users. Over the coming weeks, Google will quietly start using it "more broadly" across its services.

Get Our Best Stories!

Sign up for What's New Now to get our top stories delivered to your inbox every morning.

This newsletter may contain advertising, deals, or affiliate links. Subscribing to a newsletter indicates your consent to our Terms of Use and Privacy Policy. You may unsubscribe from the newsletters at any time.


Thanks for signing up!

Your subscription has been confirmed. Keep an eye on your inbox!

Sign up for other newsletters

TRENDING

About Matthew Humphries

Senior Editor

I started working at PCMag in November 2016, covering all areas of technology and video game news. Before that I spent nearly 15 years working at Geek.com as a writer and editor. I also spent the first six years after leaving university as a professional game designer working with Disney, Games Workshop, 20th Century Fox, and Vivendi.

Read Matthew's full bio

Read the latest from Matthew Humphries