This is the best program to enlarge images using Artificial Intelligence, and it is free

0
5
810179214 237716969 1200x630.jpg
810179214 237716969 1200x630.jpg

Artificial Intelligence is not only used to create images; It is also extremely useful for upscaling the resolution and making them larger.

Artificial Intelligence is now officially the most popular technology of 2023; From ChatGPT to Midjourney, the average user has finally discovered what AI is capable of. AI imaging, in particular, has exploded in popularity in recent months.

Projects like Dall-E or Midjourney are already capable of creating high quality images, on many occasions indistinguishable from photos or real works of art. Even glitches that normally suffered, such as people’s hands, are now rendered very realistically.

However, it is good to remember that this is only one of the things we can do with AI; There are also other tools that can be even more useful depending on our needs, and now we are going to talk about one that will become essential on any computer: Upscayl.

Upscayl is an open source project that allows you to use different algorithms to expand the resolution of our images; In other words, it makes images ‘bigger’ and sharper. The end result is as if the image had been originally created at high resolution, for example, with a higher quality camera.

Anyone who has tried to enlarge an image with an image editor such as Photoshop, Paint or Gimp will know that this It’s more complicated than it seems; The problem is the small amount of information (i.e. pixels) that a small image has. We might think that the easiest way to enlarge an image four times is simply to take a pixel and create four identical ones with the same color, and so on until completing a larger image; but the result of this process is a blurry image, which has no more detail than the original and looks worse when viewed larger.

This is where AI and the algorithms used by Upscayl come in; are able to create new details where there were none before, depending on the context of the rest of the image. The result is a new image based on the original, which looks sharper; In addition, visual artifacts can also be eliminated, such as the ‘grain’ of old photos or the ‘banding’ effect that occurs in low-quality images.

Using Upscayl is very simple. Once installed, the app shows the steps we have to take in the left part of the window. First, we must choose the image we want to enlarge, in PNG, JPG, JPEG or WEBP format (we can choose a folder by clicking on “Batch Upscayl”), and in the second step, choose the algorithm to use.

Upscayl allows you to enlarge the image up to 4 times using AI

Each algorithm has its strengths, and will produce a different result on each image; For example, if we want to enlarge a drawing or illustration, we may be interested in choosing “Digital Art”. For photographs, we have verified that “Ultramix Balanced” offers a result that is more faithful to the original image, while “Real-Esrgan” applies more effects and creates a sharper image but also more ‘artificial’; It’s a matter of taste, and we recommend trying until you find the one you like best.

Once we have chosen the destination folder, we just have to click on the Upscayl button. By default, the program applies 4x magnification to the image, although we can change this and other options in the “Settings” section. The larger the image, the longer it will take to process, something that will depend on how powerful our computer is; although Upscayl is prepared to take advantage of the power of graphics cards for this process.

Upscayl is the best image enlarging program we have seen so far, and the best thing is that it is completely free; It can be downloaded for Windows, macOS and Linux, and the source is open. Its creators are also preparing a cloud version, which will work on mobile phones and less powerful devices.

Previous articleAndroid Auto and external videos: how to send them from your mobile or USB
Next articleChatGPT: OpenAI CEO Sam Altman fired
Abraham
Expert tech and gaming writer, blending computer science expertise