Google reveals the power of machine learning

We know how important AI and machine leaning are for Google. No matter if it be a business for its cloud services or inside its devices powered by Assistant such as the Pixel phones and Home, there is room for improvement through technology, and the company is dedicated to it without doubt. And now it appears to be the turn for low res and blurry pictures.

Read More: Google Play Music Redesign Uses Machine Learning

RAISR (Rapid and Accurate Image Super-Resolution) is a prototype software

RAISR is the last magic trick out of Google’s hat. RAISR (Rapid and Accurate Image Super-Resolution) is a prototype software. Unlike traditional upsampling that enhances an image by filling it with more pixels only, RAISR makes use of machine learning algorithms to carry out the work in a more intelligent way. For example, one of its roots lays in focusing on “edge features”: when color gradients and brightness change vividly within a little time it implies that it’s the edge of a thing. Bolstering an image with the conventional way may up its resolution, however the outcome generally ends up being out of focus and blurry.

RAISR’s intelligent retouch helps to retain the original shapes more precisely, resulting in an image that just results better looking. 

In practice, at run-time RAISR selects and applies the most relevant filter from the list of learned filters to each pixel neighborhood in the low-resolution image. When these filters are applied to the lower quality image, they recreate details that are of comparable quality to the original high resolution, and offer a significant improvement to linear, bicubic, or Lanczos interpolation methods

We don’t know if we are to get a dedicated app but Google post’s conclusion appears to be hinting that it may get to us soon. 

“For example, in addition to improving digital “pinch to zoom” on your phone, one could capture, save, or transmit images at lower resolution and super-resolve on demand without any visible degradation in quality, all while utilizing less of mobile data and storage plans.”

Let us wait and see what we are actually able to get it.