Using Language to Create Imagery with GANs

EXAMPLE PROMPT: “The shadow escapes from the body like an animal we had been sheltering.” “Anatomical Art” “Anatomical sculpture” “Machine Art” “Korean contemporary painting” “Mayan Blood Letting Iconography” “color gelatin silver print”

Image Generation Prompt for VQGAN-CLIP based image generation

Using the term Artificial Intelligence or AI as a defining functionality of how these techniques work is incorrect. Generally speaking, to define the methods used to create image-based artwork using these techniques are better defined as a subset of Machine Learning: specifically, pattern matching techniques that use a rule-based methodology for assembling and blending a collage of meta-tagged images.
Although it is relatively easy to generate images with AI art using online tools, Generative Adversarial Networks or GANs accessible to Google Collab network notebooks, like with any content creation technique it’s quite difficult to create imagery that has consequential meaning. The approach I’m taking integrates the phenomena of pareidolia to infuse meaning into the imagery. With quotations by artists, poets, authors, and philosophers the imagery is complemented.

“All is visible and all elusive, all is near and can’t be touched.”
Octavio Paz

“I don’t hate them, I just feel better when they’re not around.”
Charles Bukowski

“An intellectual says a simple thing in a hard way. An artist says a hard thing in a simple way.”
Charles Bukowski

“it was an act of self-hypnosis, a deliberate drowning of consciousness by means of rhythmic noise.”
George Orwell

“I don’t hate them, I just feel better when they’re not around.”
Charles Bukowski

“I love you as certain dark things are to be loved, in secret, between the shadow and the soul.”
Pablo Neruda