Turing-ISR: AI-Powered Image Enhancements
Written by Sue Gee   
Monday, 23 May 2022

Using the power of Deep Learning, the Microsoft Turing team has built a model to improve the quality of images and is incorporating its new technology dubbed Turing Image Super Resolution in Bing Maps and the Edge browser.

We have already reported on Microsoft's Project Turing applying AI and massive compute power to Natural Language Generation, see The Third Age Of AI - Megatron-Turing NLG and now the Turing Team is tackling the problem of low quality grainy images with its new model, Turing Image Super-Resolution (T-ISR).

bing turingISR

According to Microsoft:

The ultimate mission for the Turing Super-Resolution effort is to turn any application where people view, consume or create media into an “HD” experience.  

The new technology is already available in Aerial View of Bing Maps and is currently shipped in the Canary channel of the Edge browser. It will continue to be rolled out and Microsoft must be hoping that enhanced imagery will be an attractive feature that will give its browser an edge over Chrome.

In the blog post outlining the details of the T-ISR model the Turing team explains how it adopted a two-stage process:

When we began working on super-resolution, we saw that most approaches were still using CNN architectures. Given the Microsoft Turing team’s expertise and success applying transformers in large, language models ... we experimented with transformers and found it had some compelling advantages (along with some disadvantages). Ultimately, we incorporated the positive aspects of both architectures by breaking the problem space into two stages: 1) Enhance and 2) Zoom.  

The first stage is called DeepEnhance and it was found that when testing with very noisy images like highly compressed photos or aerial photos taken from long-range satellites, transformer models did a very good job at "cleaning up" the noise, tailored to what was in the image. For example. noise around a person's face was handled differently from the noise on a highly textured photo of a forest. After using a sparse transformer scaled up to support extremely large sequence lengths and adequately process the context of an image, the result is a cleaner, crisper and much more attractive output image whose size is the same as the input.

DE turingISR

The second stage, DeepZoom, is used to scale up images and uses a novel approach. The widely used bicubic interpolation, in which pixels are inferred based on the surrounding pixel, usually results in a grainy image missing a large amount of detail that gets worse the larger the scaling factor is. To avoid this effect, the team trained a 200-layer CNN to learn the best way to recover "lost" pixel detail, The result is a model that knows the best way to recover pixels for the specific types and scenes of an image.

bicubicvturing

Using Bing Maps to look at landscapes and landmarks like St. Peter's in the Vatican City does reveal that there is a real difference in image quality thanks to the use of T-ISR. But you need to see it for yourself.

bing turingISRsq

 

More Information

Introducing Turing Image Super Resolution

Related Articles

The Third Age Of AI - Megatron-Turing NLG

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

 

Banner


Microsoft Launches Power Pages Website Builder
02/06/2022

Microsoft has announced a new way of creating business websites. Power Pages is a standalone product that is part of the wider Power Platform product family.



OpenCV Spatial AI Contest Winners Announced
17/06/2022

The OpenCV Spatial AI contest has ended with the announcement of the winners of the popular vote. OpenCV partnered with Intel and Microsoft Azure to bring an AI, computer vision, and manufacturing com [ ... ]


More News

pythondata

 



 

Comments




or email your comment to: comments@i-programmer.info

Last Updated ( Monday, 23 May 2022 )