Ain’t nobody got time to apply a filter, adjust its parameters, and tweak its lighting. Today, it’s all about text to image, and near-instant results.
As reported byPetaPixel, Facetune now lets users transform their selfies when they enter a short phrase into a text field. It’s among the latest to emerge from the explosive text-to-image AI technologies spearheaded by the likes of DALL-E, Google’s Imagen, and Midjourney.
Facetune’s ‘YOUniverse’ feature, already available within the app, combines the open-source Stable Diffusion AI model with its own smart technology to help users dream up fantasy versions of themselves in various scenes and styles—no costumes or colorful backdrops needed.
As demonstrated in the video above, users will first need to upload a few reference photos of themselves so the app can recognize their features.
From there, they will simply need to enter text prompts such as “me in the style of Van Gogh” or even more specific descriptions like “me as The Mad Hatter in Alice in Wonderland,” according to PetaPixel, before sitting back and letting the AI do its magic.
Alon David, vice president of Facetune, says the app will continue evolving with more visionary features as AI becomes more advanced. It goes without saying that the platform is on the supportive side of machine learning for art, with David noting that it makes it easier for creators to “actualize their ideas” quickly and with impressive results.