A blog related to original AI artworks. I make AI artworks using coined words and emoji as prompts.
I have been using various approaches to prompt AI Art with the name of Kandinsky, and this time, I tried combining “dividing” and “negating” the name. In other words, I provided divided versions of "Kandinsky,” such as “K_andinsky” or “Ka_ndinsky,” as positive and negative prompts for Stable Diffusion 2.1. Here are some examples of the resulting images. Although they differ from Kandinsky’s style, they are intriguing artworks.
Here are the images obtained using “Kand_insky”:
The following image is obtained from “Kandin_sky”:
Even when not specified as a negative prompt, a similar atmospheric night scene was created, as shown below:
From “Kandinsky,” the following images were obtained, which resemble Kandinsky’s paintings to some extent. In other words, various small objects are arranged together.
The face depicted on the right is vertically elongated, but as occasionally drawn by AI, the eyes are depicted as doubled.
From “Kandi_nsky,” we obtained images that resemble Kandinsky’s art to some extent, as shown below:
As for abstract paintings, there are also the following ones obtained from “Kandinsky.” While small objects are arranged similarly to the previous images, this one is considerably more chaotic.
Without further explanation, here are some additional artworks: