
As a talentless artist who likes to imagine myself as a masterful painter, I always appreciate AI tools that turn my embarrassing sketches into masterpieces.
Computer artist Glenn Marshall recently developed one of my favorite techniques yet: a text-to-image generator that transforms any picture into the style of famous artists.
The system morphs an input image towards the suggestion of a text prompt, such as “Salvador Dalí Art.” Over repeated mutations and iterations of each frame, the AI gradually finds features and shapes that match the text description until it produces a final composition.
“The results were like nothing I’ve ever seen as a computer artist for over 30 years,” Marshall told TNW. “By using any image and any text, the combination of endless possibilities is mind-shattering. And no one else is doing anything like this.”
Each piece was generated with a modified version of the Aleph-Image notebook, which is itself powered by OpenAI’s DALL-E and CLIP models.
[Read: How to use AI to better serve your customers]
Marshall named the technique Chimera, after the mythical beast formed from various animal parts, which has become a byword for something that exists only in the imagination and isn’t possible in reality.
[embedded content]
Marshall says the technique is closer to “style distortion” than style transfer. But would the artists he’s distorting appreciate his creations?
I think the surrealists would probably bow down to the wonders of AI as Gods, but the Renaissance guys would probably send the witch-hunters after me for desecrating their art with evil machines.
Published April 6, 2021 — 19:34 UTC