If you’re not familiar with Deep Image, it’s an amazing tool which allows you to increase the size of an image and upgrade its quality at the same time.

You can find it, and use for free HERE

The goal of this blog post is to focus on the main changes and showcase the results of DI 2.0 algorithms.

As we all know a picture is worth a thousand words. So we will let the enhanced pictures speak for themselves. All pictures you can see below were processed using Deep Image algorithms.

What has changed

Here are all the main improvements added to Deep Image 2.0:

  • You are now able to use a new algorithm to magnify the image two-fold and four-fold. It is based on Generative Adversarial Networks.
  • The quality of the algorithm has been improved - there are less artefacts and even smoother edges in the enhanced images.
  • We have delivered a new, more reliable asynchronous queue architecture and task processing, based on microservices.
  • You can now enjoy a fully redesigned web application

A few examples

Please keep in mind that Deep Image was trained to do a very particular job - it will increase the size of the picture as well as improve its quality once it’s enlarged. It will not improve the quality of the image you have resized before.

Check out those awesome results!

The improvements are clearly visible.

Think about all those old photos you will be able to enhance!

You are the main reason we’re working on those cool projects, so we’d love to get your feedback. Contact us and let us know what you think!