AI Art Generation Handbook/Promptcrafting
What is promptcrafting ?[edit | edit source]
Promptcrafting a.k.a Promptcraft is combination of 2 words :
Based on the words definition from Wiktionary:
Prompt : A sequence of characters that is displayed to indicate that a computer is ready to receive input
Craft : To construct, develop something (like a skilled craftsman).
It literally meant that the user / AI Artist are constructing their words skillfully to form a complete and coherent sentences that can be interpreted and
understood by a text-to-image model for it to generate the images that they wanted.
What is prompt ?[edit | edit source]
Prompt is the text input that are going to be processed by Latent Diffusion Model for it to generate images.
For successful text - to - image generation , a good prompt usually follow this format :
- Which kind of the medium / art that it is going to be ? ( The medium here is usually divided to 2 types of category: paintings and photography)
- Which artist style that you are going to follow ?
- Which type of framing techniques that you are going to use ?
- Which type of lighting techniques you are going to use ? (Notably for medium that are more to photos)
To learn more about prompts, you can head directly to chapter : Prompting in Stable Diffusion Style to understand the basis of the prompt.
Note that the prompts discussed is tested on Stable Diffusion.
Word arrangement / Word Order[edit | edit source]
Based on limited samples size, it seems that if the wording is at the front, it will have higher chances of generating the images according to your requirements. (May need more examples to correlate with this hypothesis)
As shown below, if the words of the dollar note is located at the back of sentences , it is more likely to shows Javan rhino throwing the dollar notes.
If the dollar note is at front of sentence, it is more likely to shows Javan rhino as part of dollar note designs
| Prompts | Javan rhinoceros wearing a business suit screaming aloud with hands on the cheek while seeing the stock price crash
as design on dollar note |
Dollar note showing
Javan rhinoceros wearing a business suit screaming aloud with hands on the cheek while seeing the stock price crash |
|---|---|---|
| Images |  |
 |
Modifier[edit | edit source]
Modifier in a sense is the language of AI Art Generation models and it is able to tweak the generated images into different aesthetics / according to what you are looking for.
Usually, modifier consists such as following:
(a) Art Medium
(b) Artist Style
(e) Camera Types
One or more modifiers maybe added to create the unique image generations.
Symbols[edit | edit source]
Note: This is only usable for Auto1111 and its fork (Such as SD.Next)
This symbols used is valid for use of Stable Diffusion
- Curly Braces {}: Include elements that must be present in the search results. Example: "{sunset} landscape"
- Square Brackets []: Enclose optional words or elements in the search. Example: "[Mountain] view landscape"
- Plus Sign +: Ensures the presence of a specific word in the search. Example: "Beach +{sunset}"
- Minus Sign -: Excludes specific words or elements from the search results. Example: "Forest -{animals}"
- Wildcard (*): Replaces one or more words in the search. Example: "Beautiful * landscape"
- Exclamation Mark !: Ensures the absence of a specific word in the search. Example: "Amazing {!sunset} scenery"
- Double Quotes " ": Combine multiple words into a fixed phrase. Example: "Sunset at {beach} "beautiful sky""
- Double Hyphen --: Designates a sequence of words. Example: "Hiking --mountain --{forest}"
- At Sign @: To search for results related to social media. Example: "@{city} skyline"