AI has invaded the art scene on a large scale lately. Companies like Nvidia have been working on AI tools to art from simple squiggles (opens in new tab)while robot AI artists create unique and interesting paintings. (opens in new tab) Lately, the internet is buzzing with images created by the DALL-E 2 art bot that can create images with just a few words. They are often a good mix of impressive and utterly cursed, just like this one AI Generated Duke Nukems (opens in new tab)†
While they are great for generating new mematic content on a regular basis, these AI art bots are at heart, tools and one photographer (spotted by PetaPixel) has found a good use for the software. Instead of creating an entire image from a description in DALL-E 2, they used it to clean up a captured photo and the results look more impressive than Photoshop’s.
Nicholas Sherlock (opens in new tab) is a New Zealand photographer who has used DALL-E 2 to turn some very nice but blurry images of a ladybug into sharp images in stock photography. He did this by having the AI edit his original blurry photo with the text prompt “Ladybug on a leaf, focus high resolution stacked macro photo.”
Sherlock then showed the images to the photography YouTuber Micael Widell (opens in new tab) who showed them in a video exploring the use of DALLE-2 in and possible murder in photography. This video goes through a few different examples of DALLE-2 before moving on to Sherlock’s example in photo editing, and all the results are at least somewhat terrifyingly impressive.
The video also shows another example where DALLE-2 was asked to generate other variations of a provided image and the results seem impossible to distinguish. The image generation capabilities of these AIs are staggeringly impressive. PetaPixel also compares more images of Sherlock for comparison. When rendered against images with Photoshop-generated fixes, DALLE-2 emerged as the clear winner.
The ability to flawlessly edit a real photo the way you want it seems like a great use for the tool, especially with results like this. It’s fair that in his video Widell wonders what the role of a photographer or even a hobbyist will be in a DALLE-2 AI future where anyone can create realistic images of whatever they want. Of course, there will always be other reasons to pursue these passions, but especially when it comes to things like stock photography, these AIs definitely feel like game changers.
It’s likely we’ll see an increase in AI-generated art that’s also used in games, whether it’s just to bring out concepts and explore options, or even create useful resources. There are many questions about what this means for the future of the visual arts in many industries and hobbies. However, it is important never to underestimate human desire and joy in creation. Hopefully these AIs can be a tool to amplify that more often than take it away.