Dall-E Mini is a free app that generates the image you want with AI – Tecnoblog

This week, a somewhat peculiar tool became popular on social media. Dall-E Mini is an innovative and fun web application. we are talking about a image generator that uses artificial intelligence to transform text into “abstract arts”. The best part: it’s free and available for anyone to use on their computer or smartphone.

Some attempts to create a giant iPhone in the city of São Paulo on the Dall-E Mini (Image: Reproduction)

It’s surprisingly simple to use. Just go to the app’s website, type any sentence, no matter how crazy, and press “Run”. Then, artificial intelligence will generate nine different images trying to deliver what you want. It’s not perfect, especially when it comes to human faces, but it’s certainly fun. Also, the process can take from a few seconds to a good few minutes.

AI generates any image, but it’s not very accurate

O technoblog did its own tests and the Dall-E Mini’s popularity soon became evident. It took several tries before I could really generate a demand for artificial intelligence. Each time I pressed “Run”, I received a message saying that “there is too much traffic, please try again”. Eventually, the process began.

Even though it took almost three minutes, I managed to generate a grid of somewhat distorted images of a MacBook underwater. The result is not sharp or very accurate, so it looks more like abstract generative art.:

MacBook underwater on the Dall-E Mini
MacBook underwater in Dall-E Mini (Image: Reproduction)

On Twitter, netizens have already played a lot with the artificial intelligence of the Dall-E Mini. Some results can be bizarre and funny. We have, for example:

Darth Vader playing guitar:

Kermit the Frog in a recreation of the famous The Screamfrom Much:

Gordon Ramsay eating a Big Mac (scary, isn’t it?):

Mario and Sonic at an LGBTQIA+ parade:

As you can see, some attempts do better than others. Human faces are still a challenge for artificial intelligence used in the app, but simpler things generate surprisingly good results. Landscapes, for example, can result in images that really resemble modernist paintings.

While AI-based tools of this type already existed before, this is the first to be made completely free and accessible to everyone on the internet. The site even allows the user to use it via smartphone or computer, although the result is faster on a PC.

The creator of the project is French programmer Boris Dayma, who developed the app at its current base in Houston, Texas, as part of an AI computing competition held by Hugging Face and Google in July 2021.

Google has a similar project

Google is even developing a project very similar to the Dall-E Mini. The company also uses artificial intelligence to turn any text into images. O software called Imogen is not live yetbut its developers already promise many things.

The tool’s website features some examples of how AI can create complex and surreal images, such as an oil painting of a raccoon queen wearing a red French dress. There’s also a photo of a Corgi dog riding a bike in Times Square, wearing sunglasses and a beach hat.

Images created by Google's artificial intelligence
Images created by Google’s artificial intelligence (Image: Reproduction / Google)

Unlike the Dall-E Mini, Google promises really well-finished and very accurate results. In practice, they look like professional illustrations made with 3D modeling or edited photographs.

The page also has a small demonstration, but much more limited than artificial intelligence promises. In it, you have options for image style, animal, clothes, accessories, activities and scenarios to create your image.

With information: Newsweek

Source link

About Admin

Check Also

iPhones, Alexas and Air Fryers at DISCOUNT after Black Friday

The official day of Black Friday 2022 was on the 25th of November. The date …

Leave a Reply

Your email address will not be published. Required fields are marked *