Artists use AI to generate color palettes based on text descriptions

Matt DesLauriers, an artist based in London, has created a tool that generates color palettes based on any text prompt. For example, someone could type \”beautiful sunrise\” and receive a set of colors matching a typical scene at sunset. You could also go more abstract and find colors to match \”a rainy, sad Tuesday.\”

DesLauriers posted his code to GitHub. It requires a Stable Diffusion local installation as well as Node. JS. The prototype is rough and requires some technical knowledge to install, but it shows the surprising graphical innovations which can be produced by open-source releases of powerful image synthesis model. Stable Diffusion is a neural network which generates images using tens and millions of images taken from the Internet.

Source:
https://arstechnica.com/information-technology/2022/09/artist-uses-ai-to-extract-color-palettes-from-text-descriptions/?utm_medium=social&utm_source=facebook&utm_social-type=owned&utm_brand=ars