The Sadness and Sad Eyes image was created using NightCafe AI Image Generator. IMHO NightCafe Studio is one of the best AI image generators for newbies.
Sadness And Sad Eyes
The Sadness and Sad Eyes is a simple prompt that doesn’t even mention people (no mention of girl, child, female etc) but instead concentrates on emotions, yet each try resulted in a young girl. I guess the AI assumes cute/adorable can only mean girls; I have noticed with SDXL 1.0 it tends to generate female images a lot more often than male.
If we break the AI prompt down to it’s main parts:
“An oil painting of something sad, adorable, cute and magical, with sad eyes in the style of expressive character design, Dusan Djukaric, Carne Griffiths, Lou Xaz”
We are asking the AI to produce something that’s both cute/adorable and sad/sad eyes with an expressive design. The prompt ends with three artists, “Dusan Djukaric, Carne Griffiths and Lou Xaz”
Dusan Djukaric: tends to create watercolors, landscapes, cityscapes…
Carne Griffiths: tends do create portraits, some blurb from an interview: “I like to work with an array of unusual liquids. My work is a blend of studied detail mixed with the spontaneous and chaotic. I use mainly a fountain pen and calligraphy inks which I then attack with various teas and alcohols usually thrown from the brush or sometimes with boiling water direct from the kettle.”
Lou Xaz: There’s so much AI generated art tagged with Lou Xaz I couldn’t find their original artwork!
You can see the influence of at least the first two artists in the AI images geenrated.
Ran this AI prompt with different NightCafe SDXL models, the one above uses the model RealVisXL v3 (reported to be great at faces), the one below used the default SDXL 1.0 model.
Sad Little Girl
The above was my first attempt with the prompt: “An oil painting of something adorable, cute and magical, in the style of expressive character design, Dusan Djukaric, Carne Griffiths, Lou Xaz”
I noted the expression was a little on the sad side, so future attempts included “sad” and “with sad eyes” in the prompt text.
Young Girl With Sad Eyes
The above again used the sad/sad eyes derivative of the AI text to image prompt and the default SDXL 1.0 model.
AI Image Creation Settings
Stable Diffusion XL 1.0 (SDXL 1.0).
NightCafe Creator model RealVisXL v3: An SDXL model for photorealism. Great at faces. By SG_161222.
AI Image Text Prompts
Primary Prompt
“An oil painting of something sad, adorable, cute and magical, with sad eyes in the style of expressive character design, Dusan Djukaric, Carne Griffiths, Lou Xaz”
Weight: 1
Negative Prompt
“Words, signature, ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, blurry, bad anatomy, blurred, watermark, grainy, signature, cut off, draft”
Weight: -0.3
Stable Diffusion XL has many problems with hands, feet, replicating body parts etc… added to images. This is where the negative prompt helps. By setting a negative number like -0.3 (note the negative sign -) this tells the AI to NOT do these things. So the above negative prompt is saying don’t add poorly drawn hands and face etc…
Other Prompt Settings
Initial Resolution: Medium
Runtime: Long
Seed: 2787943435 the seed number adds randomness to the AI image creation process.
Overall Prompt Weight: 70%
Refiner Weight: 5%
Sampling method: K_DPMPP_2M
