A blog related to original AI artworks. I make AI artworks using coined words and emoji as prompts.
I tried giving “QR code” as a prompt to AI art (Stable Diffusion 2.1). Of course, I didn’t expect the generated images to be readable as QR codes. While they do resemble QR codes, adding “color” doesn’t actually add colors, and adding “3D” does not make them three-dimensional. So, when I specified “QR code” as both positive and negative prompts, I obtained various images with colors and placed in a 3D space.
First, let’s see the results when only “QR code” was specified as a positive prompt.
The image on the right has a little bit of color, but out of 12 generated images, only this one has colors. Although some images were generated with “3D” specified, they didn’t actually become three-dimensional.
In contrast, when “QR code” was specified as both positive and negative prompts, I obtained images with various shapes and colors, including some placed in a 3D space.
Both of these two images evoke an animalistic presence. They could be interpreted as resembling a barn owl or something similar.
In the image on the top left, it appears as if a three-dimensional “QR code” is adorned on the walls of a room. Furthermore, in the image on the top right, a “QR code” is layered onto a landscape with a plateau. The pattern of the foreground soil bears resemblance to a QR code, suggesting a connection to the layering process.
Finally, I will include two more images that showcase the colorful paintings resulting from specifying negative prompts.