The Text-to-Image AI Generator: the Revolution in Imagery Production

Just think about a text description: the Text-to-Image AI generator converts those words into images. As one of the massive trends in the machine learning sector, it’s predicted to be the start of a revolution in imagery production (they do, in fact, eliminate the need for technical labor in the creation of pics).

Alessia Gaspodini   ·   Sep 6 2022

Just think about a text description: the Text-to-Image AI generator converts those words into images. As one of the massive trends in the machine learning sector, it’s predicted to be the start of a revolution in imagery production (they do, in fact, eliminate the need for technical labor in the creation of pics).

In January 2021, OpenAI, an artificial intelligence research laboratory, released under the name of DALL-E their new text to image generator AI and, after only one year, they had already developed a new system, DALL-E 2, which has a 4x greater resolution. You might have seen an astronaut riding a horse in a photorealistic style, or a painting of a fox sitting in a field at sunrise in the style of Claude Monet.  

an astronaut riding a horse in a photorealistic style,    a painting of a fox sitting in a field at sunrise in the style of Claude Monet.


What you see here are 2 of the results of this new text-to-image generator AI, which recently hit the web enormously. This innovation can also edit existing images, add and remove elements, tones, and shades. It can also create different variations of an existing image. Ready to have your unique images whenever you want? During summer 2022, OpenAI announced that DALL-E is now accessible as a public beta (meaning that the service is available for any member to use). You can play around with the existing images and combinations on the DALL-e 2 website, you can choose your main characters (i.e., teddy bears) and see how the AI locates them within different contexts.

Here are teddy bears working on new AI research on the moon in the 1980s.

teddy bears working on new AI research on the moon in the 1980s

While here are teddy bears working on new AI research underwater with 1990s technology.

teddy bears working on new AI research underwater with 1990s technology.

Besides this website, there are also some other tools that you can use to play around with AI and create your own projects. Starry AI, Dream by Wombo, and Nightcafe are some of the websites you can check out to make room for your imagination. (Yes, you should try it out. Research shows that boosting your imagination and creativity enhances problem solving skills, develops your self-confidence, and even saves you money). Go ahead and trigger your curiosity!

What would you ask this machine to create? What would you give life to? You're one click away to become an AI artist!

Alessia Gaspodini-rounded

"For example, here’s my text-to-image AI result. I love travelling and I love flowers." I got travelling flowers drawn.

travelling flowers 

Conclusion
Because AI text-to-image technology is revolutionizing the creation of imagery, it also plays a significant role in the world of NFTs. For example, StarryAI, the art generation tool that has been mentioned before, allows you to turn your own projects into unique NFTs. As of August 15, 2022, on the Ethereum, Ronin, and Flow blockchains, there had been registered a total of about 21.4 thousand sales of NFTs in just 30 days (source: Statista). But this is not surprising if we think that around 360,000 people own NFTs and only 32,400 own an estimation of 2.7 million distinct NFTs.

Fun Fact: The first NFT ever was created in 2014 by Kevin McCoy and Anil Dash, and it consists of a video clip made by McCoy's wife, Jennifer.

And you? Do you have NFTs?

 

Hope you had fun,
Alessia from MAPEGY

Alessia Gaspodini