Man simulates time travel thanks to Stable Diffusion image synthesis

All through December, a social media user acknowledged as Stelfie the Time Traveller has been crafting a time-hopping travelogue making use of generative AI. Many thanks to Steady Diffusion and fantastic-tuning, an nameless artist has established a fictional photorealistic character that he can insert into fake historical photographs set in distinctive eras, this sort of as historical Egypt or the time of the dinosaurs.

Steady Diffusion is a deep studying impression synthesis product that allows people today to create fictional scenes using text descriptions termed prompts. With an supplemental system known as Dreambooth, folks can insert their possess subject or character into scenes produced by Secure Diffusion. It can also be applied to insert real people today into fictional cases.

So much, “Stelfie” has taken historic selfies for the duration of the ice age (getting chased by a woolly mammoth), in historic Egypt (for the duration of the design of the pyramids), in historic Greece (with the Trojan Horse), hanging out with Leonardo da Vinci (when generating the Mona Lisa), in the old West, whilst running from a tyrannosaurus rex, and whilst sailing with Christopher Columbus.

The artist driving Stelfie writes social media posts (on Twitter, Instagram, and Reddit) in character as playful dispatches from a 41-yr-previous time traveler as he visits unique locations, for instance:

Log #3 – Columbus ship, 1492 DC. “Laaaaand” Now, I you should not wanna brag but… Just after months of sailing Columbus was shed in the Oceans and very well let’s say I may have kinda suggested the right direction…

The anonymous artist (a self-identified “humorous outdated gentleman”) in-depth some of the course of action he utilizes to produce the photos in several Reddit opinions: a mix of Steady Diffusion 1.5, a personalized AI product for the landscape, and a customized AI design skilled on the Stelfie face, which is apparently a fictional person developed working with Character Creator. He takes advantage of “a ton of inpainting,” which implies inserting AI-produced imagery into the photographs to correct errors and sculpt the scene, and each and every image normally takes a few several hours to create.

Some might locate Stelfie’s makes an attempt to provide the photographs as NFTs unpalatable (the tech is controversial), but simulated time travel is arguably a novel entertainment use of engineering that can normally generate deepfakes and perhaps deceive others. Using Dreambooth and Secure Diffusion, any individual can now insert a character (or themselves) into great and fictional conditions, so we’ll probable see a large amount much more fictional leisure situations in the near foreseeable future.

Listing image by StelfieTT