How To Write AI Image Prompts: Guide for Creators

Artificial intelligence (AI) image generation lets non-artists make professional-quality visuals in seconds. To create the best AI art, you’ll need to understand how to craft effective AI image prompts. Writing with proper structure and sufficient details helps you generate any picture you can imagine. 

Keep reading to learn how to make AI photos and write image prompts.

AI Image Prompt Structure

An AI image prompt instructs a generative AI model to create pictures. Clear descriptions explain the desired visual in detail, helping the model understand the user’s intent and create a picture that matches their vision. 

While you can construct AI art prompts however you like, strong inputs tend to use a similar structure that contains most or all of the following criteria:

• The subject description refers to the image’s focal point. It’s the person, place, animal, object, or idea performing the action in the image. 
• Style and medium specify the form or art style used in the visual, such as illustration or photograph. You can reference artists or artworks, like “in the style of Salvador Dali’s Persistence of Memory,” or specific techniques, like “using pointillism in the Neo-Impressionist style.”
• The background scene explains where the action performed in the image occurs, such as a dog park or outer space. 
• Composition elements cover all the additional aspects of the artwork, including the image’s aspect ratio, perspective, and color scheme.

How To Write an AI Image Prompt in 4 Steps

Once you know how to structure an image prompt, you can explore how the AI interprets instructions by adding specific details. Follow these steps to generate reliable results — once you’ve mastered the basic formula, experiment with different structures, lengths, and styles.

1. Define the Subject Clearly

Begin your image prompt with a clear description of your subject. This covers what’s happening in the image, such as “a family on a picnic” or “a cat sleeping on a chair.” Depending on the context, you may want to add the following details:

• Actions performed by the subject — Walking, eating, reading
• How the subject performs actions — Quickly, lazily, loudly
• The time of day — Morning, noon, night
• The background — A forest, an apartment, an alleyway
• The mood — Chaotic, tranquil, exciting

2. Indicate the Type of Image

You may want your image to have a specific aesthetic or imitate works by a famous artist. To achieve this effect, incorporate these common styles in your prompts:

• Photography — Mention the camera and lens type, exposure, and saturation.
• Illustration — Describe the medium, such as pencil, charcoal, and cartoon.
• Digital art — Specify styles like pixel and fractal.
• Sculpture — Explain whether the image should have a relief, carved, or cast appearance.
• Animation — List aesthetics like 3D and abstract, or mention studios like Disney and Ghibli.
• Painting — Mention techniques (brushwork and texture) and movements (art nouveau and cubism). 
• Artist names — List the artist whose style you want to emulate, such as da Vinci or Frida Kahlo.
  

3. Provide Additional Detail

The more detail in your text, the better your results. When writing prompts for AI image generators, structure descriptions by priority. Start with the most essential details, and follow up with secondary elements that add context and refine the final look:

• Color scheme
• Framing
• Lighting
• Perspective
• Mood
• Angles
  

4. Clarify With Negative Prompts

Use negative prompts at the end of your description to avoid unwanted elements. For example, adding “no glasses” or “no background text” excludes those items from the image.

Best AI Image Prompt Strategies

For some images, you may need to use more advanced prompting strategies like the following to achieve your desired result:

• Image-to-image — Rather than writing your instructions, try using an image to prompt the generative AI model to create your picture. This technique, known as image-to-image prompting, involves providing the AI tool with a photo it can use as inspiration.
• AI outpainting — AI outpainting is a technique for expanding an image beyond its original frame. To use this method, upload a photo and watch as AI creates a new, larger piece. 
• Reverse engineer images — This process works in reverse: Instead of using text to generate an image, you upload a real photo to create a description. This technique shows how AI interprets visuals and helps you write better prompts.
• Adjust generation settings — Some AI models, such as Stable Diffusion, let you adjust the number of stages the tool goes through when generating an image. Adding more steps typically produces better results.
• Use AI to enhance your prompts — If you’ve got writer’s block, use AI to create image prompts for you. For instance, turn on Captions’ Magic Prompt tool so AI polishes your rough descriptions.

AI-Created Image Prompt Examples

If you need some inspiration, check out these prompt examples that you can adapt to your own content:

• Sci-fi and fantasy — An astronaut in a futuristic space suit running across a rocky alien planet, plants on the ground glowing softly, high-definition details, in the style of Ron Walotsky.
• Art portrait — A colorful portrait of a fuzzy kitten, multicolor with high color contrast, high saturation, vector art style, no harsh shadows.
• Landscape — A peaceful Japanese Zen garden in spring, cherry blossoms falling gently, a red arched bridge over a koi pond, realistic texture, no harsh lighting, and no unnatural colors. 
• Product — A realistic high-angle product photo shot of a red sports car driving along a winding canyon road at dusk, with pine forests and mountains in the background, natural lighting, and no distorted features. 
• Illustration — An illustration of dinosaurs dancing around a campfire, bright pastel colors, and simple lines, in the style of artists like Sandra Boynton and Axel Scheffler. 

How To Generate an Image With Captions

Captions’ built-in integrations let you create images with AI models like Luma and DALL-E, all within a single dashboard. Here’s how it works:

• Upload a new project — Import your footage to Captions.
• Customize your settings — Choose “Image” as the media type, and pick an AI model.
• Input your text — Describe your picture using the structure and strategies laid out above. Toggle on “Magic Prompt” to have AI refine your writing.
• Adjust the output — Optionally, set the aspect ratio.
• Tap “Generate” — Create your image, and hit the “Regenerate” image to make a different version.
• Customize — Add the picture to your video, and adjust how long it’s on screen, where it appears, and how big it is.

By following these simple steps, you can easily make great videos with pictures.

Take Your AI Images to the Next Level With Captions

With AI image prompts, you can kickstart the image creation process, saving you time and money. If you want to create captivating images to include in your content, look no further than Captions.

Captions is an all-in-one AI video editing platform that comes equipped with integrations for generative AI tools like Luma AI and DALL-E. Create visuals for your content in seconds, and while you’re at it, add captions, royalty-free music, or AI avatars — no experience required. 

Give Captions a try and boost your content with realistic AI images.