midjourney keyword frame mapping

The effect of Midjourney’s pictures is always not what I want

{Why is this? See below for solutions)

Is there anyone like me? ?Jian just copied and pasted other people’s keywords at first, and then changed them and produced the picture. The final effect was unsatisfactory! Putting aside the randomness of Midjourney's pictures, another reason is that we don't have enough control over keywords. In fact, many keywords and grammar are of little use to Midiourney's generation logic (invalid keywords, grammar ), so some images that do not meet expectations are often generated. In fact, the official has given a standardized grammatical structure

#01.

What is Prompt?

"Prompt" refers to the short text phrase used to generate the image. Midjourney Bot breaks the words and phrases in the prompt into smaller units called tokens, which can be compared to its training data and then used to generate an image. This means that no matter what the sentence pattern is, MJ will first disassemble the long sentence, and then generate the final picture through the "black box algorithm" of multiple tokens. Therefore, the meaning of language order is not significant.

#02

MJ doesn’t understand grammar at all

Midjourney does not understand grammar, sentence structure, or words like humans do. In many cases, more specific synonyms are more effective. For example, you can use words such as "huge" and "huge" to replace "big". Use fewer words where possible. Using fewer words means each word has a more powerful impact. Use commas, parentheses, and hyphens to help organize your ideas, but be aware that Midiourney will not reliably interpret them. Midjourney Bot is not case sensitive.

#O3

MJ official Prompt recommendation

A simple Prompt may be just a word (or even an emoticon), which can generate a picture. A very short prompt will rely heavily on Midiourney's default styling, so a more descriptive prompt will produce a more unique effect. However, prompts that are too long are not very good.

Please try to be as clear as possible in describing any context or details that are important to you. Consider the following factors: theme, style, environment, lighting, color, mood, composition

#04.MJ official prompt recommendation

Theme +

Characters, Animals, characters, objects, etc.

Style+

Photos, paintings, illustrations, sculptures, graffiti, tapestries, etc.

Environment+

Indoor, outdoor, lunar, Underwater, Emerald City, etc.

Lighting+

Soft, ambient, cloudy, neon lights, studio lights, etc.

Color+

Vivid, soft, bright, monochrome, colorful, black and white, etc.

Mood+

Calm, peaceful, noisy, energetic, etc.

Composition

Portraits, avatars, close-ups, aerial views, etc.

#05

MJ official Prompt recommendation

Plural forms of words may bring a lot of randomness.

Please try to use specific numbers. For example, "three cats" is more specific than "cat"

Finally, I would like to share with you a Midjourney dictionary produced by [Tai Kelang]. All keywords have been tested by the team. ***Calculate thousands of common vocabulary, and carefully screened 50 practical keywords, including style, lighting effects, material performance and other elements