Stable Diffusion Eye Color Prompt

Stable Diffusion is an AI image generation model that allows users to create realistic images through text prompts. One common use case is generating portraits with customized eye colors. Crafting effective prompts for modifying eye color can be challenging, but following some key best practices can help.

In this article, we will provide AI prompt examples and techniques for successfully changing eye color in Stable Diffusion portraits.

Prerequisites

Before diving into eye color prompts, it’s helpful to understand some basics:

  • Stable Diffusion is trained on image-text pairs, so prompts need to describe the desired visual output
  • More details and context in prompts typically produce better results
  • Using clear language and avoiding ambiguity helps the AI understand
  • Prompts can leverage special keywords that control image generation

With these fundamentals covered, let’s look at prompt engineering strategies for eye color.

Set Up the Base Prompt

First, establish the foundation of your portrait prompt without any eye color specifics:

A realistic portrait of a smiling woman with long blonde hair wearing a red sweater, by Greg Rutkowski

This sets the overall scene. Now we can add eye color details.

Use a Separator

Adding a separator word tells Stable Diffusion to treat text before and after differently:

A realistic portrait of a smiling woman with long blonde hair wearing a red sweater, by Greg Rutkowski, with

Common separators include with, has, and break.

Specify the Eye Color

Next, define the exact eye color wanted:

A realistic portrait of a smiling woman with long blonde hair wearing a red sweater, by Greg Rutkowski, with bright green eyes

Use descriptive and unambiguous language here.

Adjust Intensity

Including intensity modifiers lets you control how prominently the eye color appears:

A realistic portrait of a smiling woman with long blonde hair wearing a red sweater, by Greg Rutkowski, with very bright green eyes

Try terms like very, piercing, deep, etc.

Add More Context

Extra context about lighting, reflections, and other details can enhance realism:

A realistic close-up portrait of a smiling woman with long blonde hair wearing a red sweater, by Greg Rutkowski, with very bright green eyes reflecting soft window light

Use Multiple Methods

For best results, combine several prompting techniques. For example:

A beautiful young woman by Greg Rutkowski, with vibrant green eyes glistening in the sunlight. The green eyes stand out clearly against her fair complexion.

break

Woman with striking emerald green eyes, soft skin, rosy cheeks, blonde hair. Green eyes reflect warm sunlight. Photorealistic portrait by Greg Rutkowski.

This leverages separation, intensity modifiers, context, and duplication across two prompts.

Advanced Prompting Methods

Beyond basics like intensity words and separators, there are more advanced tactics for fine-tuning eye color.

Adjust Color Tones

Include color tone descriptors to shift the eye color hue:

Woman with very bright lime green eyes, soft skin, rosy cheeks, blonde hair. Green eyes reflect warm sunlight. Photorealistic portrait by Greg Rutkowski.

Try lime, forest, teal, emerald, etc.

Use Comparisons

Compare the eye color to physical objects with similar hues:

Woman with eyes the color of ripe limes, soft skin, rosy cheeks, blonde hair. Green eyes reflect warm sunlight. Photorealistic portrait by Greg Rutkowski.

Add Stylistic and Lighting Details

Extra context about lighting, reflections, contrast and other stylistic factors helps the AI render the eyes realistically:

Close-up portrait of a young woman with eyes the vibrant green of ripe limes. Her eyes glimmer in the soft sunlight. Long blonde hair and rosy cheeks contrast beautifully with her fair complexion and green eyes. Photorealistic painting by Greg Rutkowski. 

Prompt Structure Tips

Keep these structural recommendations in mind when crafting eye color prompts:

Use Clear Separators

As shown in the examples above, separators like with and break indicate transitions. This clarifies which text applies to which image aspects.

Put Eye Color Late

Save eye color terminology for the end of the prompt. This prevents it from being overridden by other prompt elements.

Limit Prompt Length

Concise prompts are best. Try to keep prompts under 300 characters including spaces.

Avoid Ambiguity

Be as unambiguous as possible – say green eyes rather than just green.

Use Multiple Prompts

Generating several variations of a prompt can produce better results through iteration.

Common Eye Descriptors

Here is a handy list of eye color descriptors to work into prompts:

Intensity: vibrant, piercing, bright, vivid, striking, bold, dazzling, gleaming

Color Tones: emerald, lime, forest, jade, teal, cyan, gold, yellow, purple, red

Stylistics: glimmering, sparkling, shining, shimmering, glistening, gleaming, glittering

Lighting: reflecting, with catchlights, illuminated

Adjectives: jewel-toned, rich, deep, light, pale, bright

Prompt Recipe Template

Use this madlib formula to easily craft eye color prompts:

A beautiful close-up portrait of a young [ethnicity] woman with [base colors] [base features]. [Separation word] eyes the color of [descriptive color] [gemstones/objects] that [stylistics] in the [lighting]. [Additional context]. Photorealistic painting by [artist].

Fill in the brackets with appropriate terms and descriptors to generate targeted, effective eye color prompts.

Conclusion

Modifying eye color in AI-generated portraits requires thoughtful prompt engineering. Following best practices around structure, specificity, descriptors and context sets up prompts for success.

Leveraging the recommendations in this guide will help you master eye color prompts with Stable Diffusion. Feel free to build on these examples and techniques to create stunning portraits showcasing dazzling eyes of any hue.