What if you could write a description of a painting you would like to see, and then in a matter of seconds a computer will generate that specific image? DALL-E-2 does exactly that.
DALL-E-2 is a new AI system developed by OpenAI that can create realistic images and art from a description in natural language.
DALL-E-2 has an option to upload your own images to the system and it generates several versions of its own for you. The program reads the image and produces versions of it in several different styles, positions and more.
DALL-E-2 has learned the relationship between images and the text used to descirbe them. It uses a process called “diffusion” which starts with a pattern of random dots and gradually alters that pattern towards an image when it recognizes specific aspects of that image.
There was a big improvement compared to the previous system (DALL-E) – the new system generates more realistic and more accurate images with 4* greater resolution.
In order to use DALL-E-2 there’s an option to enter your details in their website and get on their waitlist. It takes a couple of months to get in but its probally worth it
Heres an example of an image i described to DALL-E-2: “Satan trying to sell a used tesla to jesus in a car dealership in the style of gustave dore etching”
As you can see, the details are not at all specific and their faces are blurred, but there’s no doubt that the result is amazing:
