The excellent Pixelmator Pro image editor for the Mac was updated today with a cool new feature allowing you to boost the image resolution while keep details sharp, without introducing visible image artifacts like pixelation or blurriness.
They’re calling the feature ML Super Resolution and it takes advantage of machine learning.
Developers note in a blog post that zooming and enhancing images like they do in all those cheesy police dramas “is now a reality”. Specifically, they claim scaling up an image to three times its original resolution without visible artifacts.
As computers get ever more powerful, the additional power opens up new possibilities. One of the uses of machine learning, on a very fundamental level, is to make predictions about things.
In this case, we gathered a set of images, scaled them down and then ‘taught’ the algorithm to go from the scaled-down version to the original resolution, high-quality image, predicting the values of each new pixel.
The algorithm can’t recreate detail that is too small to be visible but it can make amazing predictions about edges, shapes, contours and patterns that traditional algorithms simply cannot.
This is all quite taxing on the CPU and GPU.
On older Macs, it can take minutes to process a single image. On the latest hardware, images are processing in a few seconds, developer claim, and even faster on machines such as the iMac Pro, Mac Pro or any Mac with multiple GPUs. Because this is a hardware-accelerated feature, ML Super Resolution is also significantly improved when using an eGPU.
You’re encouraged to visit the official blog post for some cool interactive examples of what ML Super Resolution can do for you. Basically, their visual examples compare the new ML-powered algorithm to the usual image scaling algorithms — Bilinear, Lanczos and Nearest Neighbor — found in the app’s Image Size tool.