Imagine if you could create a photo, any image at all, with just a few words and, no matter how bizarre or nonsensical your concatenation might be, an AI somewhere is capable of generating output based upon what you input.
That’s pretty much the proposition behind the DALL-E 2 photo generator and, while we’ve seen other AI photo-generating software out there before, we haven’t seen one quite this capable.
You can view some examples of the DALL-E 2 in action over on the website at this link.
A description of the process, while succinct, leaves out a lot of the magic:
“DALL·E 2 can make realistic edits to existing images from a natural language caption. It can add and remove elements while taking shadows, reflections, and textures into account.”
You can also take images that already exist and have DALL-E apply a style to them, such as “in the style of Andy Warhol” and “as a pencil drawing.”
How does it do this?
Simple, at least according to the website showcasing it all: “DALL·E 2 has learned the relationship between images and the text used to describe 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.”
Currently, DALL-E 2 is part of a research project and isn’t monetized as a result although you can join a preview waitlist over on their website to stay up to date on things if that changes in the future.
What uses could you see the Dall-E going toward in the future? Let us know your thoughts in the comments below.
We’ve got some other photography news for you to check out at this link right here.