<aside> 💬
According to the website, CLIP Interrogator 2 writes a “good prompt to create new images like an existing one”. The specific version linked above and used in this project is “specialized for producing nice prompts for use with Stable Diffusion 2.0 using the ViT-H-14 OpenCLIP model”.
This software was used because it is accessible, easy to use, completely free with unlimited use, and has no prompt word limit—getting as detailed as possible, when the settings are tuned.
All images were processed in “best mode” and max flavors (24). After 20-60 seconds, a detailed description is ready for usage in image generation software.
</aside>
<aside> 🖼️
According to their About page, OpenArt is a “platform dedicate to AI art enthusiasts”. The website employs Stable Diffusion in its AI art creation.
This software was used because there is no word limit for its prompts. Though it is primarily a paid software, I was able to garner trial credits by joining their Discord community.
OpenArt also allows for a combination of words and image in its prompts, which is crucial for this project.
All images were processed with a combination of detailed word prompts and image prompt.
A snapshot of my creation history, showing all the images in this project:
</aside>