Recent OpenAI release of the CLIP model unleashed using it for free-form text-to-image synthesis methods in the wild. The novel tool of such kind, developed for this submission, “translates” text to the massive layered textures, resembling both medieval paintings and legendary “deepdream” on a new level, with a meaning and context.
The terma Aphantasia means the inability to visualize mental images, the deprivation of visual dreams. May such tool be an antidote for that? Think an outsourced visual imagination, triggered by words..
Associated artwork is a cinematic rendering from the tool. It has processed the titles of a few iconic paintings, novels and music classics, interpolating through the list of subjects. Essentially, this set a new kind of reproduction: as a reconstructed story, an AI-learned view on topic. These distilled visuals may be seen as modern memes, distorted casts of the centuries-old narratives, unfolded from their short names.
Welcome to spell your own casts and summon the visions: github.com/eps696/aphantasia
Sound for video: Archon Orchestra “Saltarello” Back To The Gallery