Lecture

Image Prompt Structure to Minimize Trial and Error

To minimize trial and error in creating the desired image, it is advisable to include the following five components in your prompt.

Basic Structure of Image Generation Prompt
[Modifier] + [Subject] + [Detailed Description] + [Style] + [Aspect Ratio]

Modifier & Subject: Use adjectives or emotions as modifiers to describe the subject "specifically"

  • Example: Puppy → Lovely with curious eyes, newborn Golden Retriever

Detailed Description: Describe the background of the image, express lighting effects, or specify details such as image quality

  • Example: Calm photo → Calm 4K quality image with warm afternoon sunlight from the top right

Style: Determine the type and overall feeling of the image

  • Example: Santa Claus → Santa Claus on a sleigh pulled by two reindeer, in a post-impressionist Van Gogh style

Aspect Ratio: Specify the aspect ratio of the generated image according to its purpose

  • 1:1(Square), 7:4(Landscape), 4:7(Portrait)


Prompt Example

Image Generation Prompt Example Using Basic Structure
A futuristic city with a cyberpunk feel filled with neon signs, a 4K quality image capturing the cityscape under a dark night sky, with a professional photo-like feel, aspect ratio of the generated image is 7:4
StructurePrompt
Modifier & SubjectA futuristic city with a cyberpunk feel filled with neon signs
Detailed DescriptionA 4K quality image capturing the cityscape under a dark night sky
StyleWith a professional photo-like feel
Aspect RatioAspect ratio of the generated image is 7:4
Mission
0 / 1

What is the most suitable word to fill in the blank?

The basic structure of an image generation prompt is ' + subject + detailed description + style + aspect ratio'.
Modifier
Subject
Detailed Description
Style

Lecture

AI Tutor

Publish

Design

Upload

Notes

Favorites

Help