Google Releases 'Hollywood-esque' Low-Res Image Enhancement Software

Google Releases 'Hollywood-esque' Low-Res Image Enhancement Software

We’ve all been there. One of our images is slightly out of focus, or one of our friends has fired over a super low-res image and, as their neighborhood photographer, asked if there is anything we can do to salvage it. Thanks to Google’s new image resolution enhancement software RAISR - which is drawing comparisons to the "magical" software we often see in TV and film - we may be in with a chance.

 

The software works by incorporating "machine learning" in order to significantly increase image resolution. Google even go as far as to claim that the software uses 75% less bandwidth per image.

 

In the words of Google Research Scientist Peyman Milanfar:

RAISR produces results that are comparable to or better than the currently available super-resolution methods, and does so roughly 10 to 100 times faster, allowing it to be run on a typical mobile device in real-time. Furthermore, our technique is able to avoid recreating the aliasing artifacts that may exist in the lower resolution image.

The aforementioned "machine learning" is a technique that uses both high and low quality images to find filters that are selectively applied to each pixel of the low resolution image. 

Check out the video below from PiXimperfect for an in-depth explanation. You can go ahead and skip to 2:40 for the good bit!

https://www.youtube.com/watch?v=wQhLNByOMyk

It’s still early days for the prototype, but is an exciting development nonetheless, and one that can hopefully bring restoration of low-res images can be available to the masses in the coming years.

 

[via SLRlounge and Google Blog]

 

Jack Alexander's picture

A 28-year-old self-taught photographer, Jack Alexander specialises in intimate portraits with musicians, actors, and models.

Log in or register to post comments
15 Comments

Literally the first thing that came to my head when I read this.

It's an interesting method of image data compression but I'd think it's true value would be in video applications.

I'm not a photographer, for me photography is just a hobby. But as I see this kind of technology improving I see a potential risk for people making a living from selling their images, because with such a technology I can "steal" the low res photo from a photographer's website and make myself a higher resolution one, print it or resell it.

I was thinking the same... I purposely upload low res images only.

Another thing to worry about ...

It's a little early for April 1st stories, but I guess I can sell my 50MP camera now ;)

This is disturbing on so many levels. While it sounds wonderful, the results may be that we can never upload anything again (even at the lowest resolution) without the possibility of it being downloaded ('err stolen) , enhanced with something like this, and wham, we have lost control of our image forever.

Getty will be very happy about this. They love stealing images.

Only one brief blog entry from google on this and it focusses on bendwidth reduction for G+

I am on the fence regarding this as the only images shown on-line and in videos are small thumbnails, so you cannot get a true sense of its performance. The old saying is that you cant make something out of nothing,,,

Where have they "released" it?

It is incorporated in Google Plus. It is meant to make viewing of large photo file faster and save on bandwidth

I don't buy it. I've seen this advertised before, and the samples I've seen do not look that great.

My understanding of RAISR is that it makes viewing photos in Google Plus faster, not sure if it is capable in doing things that are mention here. That is what it is when it was first announced back late last year.

Pfff I get better results with Photolemur.com