This prompt turns a single reference photo into a photorealistic three-frame fashion collage while keeping the person's original identity intact. The emphasis is on consistent facial features, natural body proportions, realistic clothing, and a soft outdoor editorial look rather than an overly polished AI-generated appearance.
Get This Prompt
The Prompt
Create a photorealistic 3:4 vertical fashion collage divided into three horizontal frames with soft, seamless transitions between each section.
Keep the person's original face and identity exactly the same as the reference image. Preserve the facial features, face shape, skin tone, hairstyle, eyes, nose, lips, and natural skin texture. Do not alter, beautify, or replace the face.
Show the same person wearing a stylish oversized shirt with dark pants, captured in three different natural and emotional poses. Maintain consistent appearance, outfit, body proportions, and identity across all three frames.
Frame the subject according to the available screen space. Keep the composition balanced, leave natural breathing room around the subject, avoid cutting off the head or body, and ensure the main subject remains clearly visible in every frame. Use a clean portrait composition optimized for a 3:4 vertical display.
Background: soft blurred green grass fields, distant trees, warm natural light, dreamy bokeh, and shallow depth of field.
Soft emotional mood, cinematic fashion aesthetic, natural human proportions, realistic fabric texture, authentic skin details, subtle colors, realistic lighting, editorial fashion photography, highly photorealistic.
Avoid excessive AI beauty effects, plastic skin, distorted facial features, unnatural body proportions, harsh sharpening, or an artificial CGI look.Negative Prompt
plastic skin, altered facial identity, different face, beautified face, distorted facial features, inconsistent face between frames, unnatural body proportions, extra limbs, malformed hands, duplicate person, mismatched clothing, cropped head, cropped body, harsh sharpening, CGI appearanceRecommended Settings
| Model | Gemini |
| Aspect Ratio | 3:4 vertical |
| Reference Image | Use one clear photo of the person |
Good to Know
Face preservation is an instruction, not a mathematical guarantee. Results can still vary depending on the reference image, the model version, and how much visual change the edit requires.
Table of Contents
What This Fashion Collage Prompt Creates
The prompt is designed around one specific challenge: turning a single person's reference image into a three-frame fashion composition without making the three versions look like different people. Each frame has its own pose, but the face, hairstyle, outfit, body proportions, and overall identity are explicitly tied together.
The visual direction is intentionally soft. Green grass, distant trees, warm natural light, shallow depth of field, and dreamy bokeh create an outdoor editorial setting without competing with the subject. The 3:4 vertical format also gives the composition enough height for fashion poses while keeping the subject suitable for portrait-oriented screens and social content.
Why Gemini Fits This Prompt
Gemini is the strongest fit when preserving a specific person's recognizable identity is the main requirement. This prompt is not simply asking for a new fashion model. It starts with a reference image and asks the model to retain facial features while changing the visual presentation around that person.
That distinction matters. A highly stylized image model may produce a beautiful fashion portrait but still reinterpret the person's face. Here, facial identity is more important than artistic reinterpretation, so an instruction-focused editing workflow makes more sense.
For this use case, the important control is the instruction itself rather than a large collection of technical generation parameters. Gemini's image workflow uses aspect ratio as a relevant setting, while the prompt carries most of the visual direction.
How to Use the Prompt
- Upload one clear reference photo of the person you want to preserve.
- Use the complete prompt above and keep the face-preservation instructions intact.
- Set the output to a 3:4 vertical composition when the image workflow provides an aspect-ratio option.
- Check all three frames for facial consistency, body proportions, clothing continuity, and accidental cropping.
- If one frame looks noticeably different, refine that specific issue instead of adding a long list of unrelated visual adjectives.
Why the Prompt Uses Three Horizontal Frames
The three-frame structure gives the final image the feel of a fashion editorial contact sheet while remaining a single vertical composition. Each section can show a different emotional pose, but the repeated identity and clothing create continuity between them.
The instruction to use “soft, seamless transitions” is important because hard borders would make the result look more like three unrelated photographs placed next to each other. The transitions instead allow the backgrounds and lighting to visually connect the frames.
Keeping the Face Consistent
The strongest part of the prompt is the explicit identity-preservation section. It does not simply say “keep the face the same.” It names the features that matter: face shape, skin tone, hairstyle, eyes, nose, lips, and natural skin texture.
That specificity helps establish what should remain stable while other elements are allowed to change. The prompt also tells the model not to beautify or replace the face, which is particularly useful for avoiding the common result where an edited portrait gradually becomes a generic AI fashion model.
Reference Image Quality Matters
Choose a reference where the face is clearly visible and not heavily filtered. A front-facing or moderately angled portrait with natural lighting generally provides more useful facial information than a distant photograph with strong shadows.
Do not deliberately choose a low-quality reference and expect the prompt to restore every facial detail perfectly. The source image still establishes much of the visual information available to the editing model.
Customizing the Fashion Look
The current prompt uses an oversized shirt and dark pants because the combination keeps the attention on the subject while giving the three poses enough visual consistency. You can change the outfit without changing the underlying structure of the prompt.
For example, replace the oversized shirt with a relaxed linen shirt for a summer editorial look, a structured jacket for a sharper fashion composition, or a simple neutral sweatshirt for a casual street-style direction. Keep the clothing description consistent across all three frames if character continuity is important.
The background can also move away from the green field. A quiet urban street, beach at golden hour, minimalist architectural location, or autumn park can create a completely different editorial mood while retaining the same three-frame composition.
Lighting and Mood
The prompt combines warm natural light with shallow depth of field and dreamy bokeh. These choices work together: the warm light establishes the emotional tone, while the shallow depth of field separates the person from the grass and distant trees.
If you want a more dramatic result, replace the warm natural-light direction with soft overcast lighting or late-evening golden light. Avoid stacking several contradictory lighting instructions. Asking for soft daylight, harsh rim light, studio lighting, and dramatic shadows simultaneously gives the model competing priorities.
Common Mistakes to Avoid
- Using a weak reference image: A blurry or heavily filtered face makes identity preservation less reliable.
- Changing too many things at once: Major changes to face, body, clothing, environment, lighting, and pose increase the chance of inconsistency.
- Ignoring the three-frame relationship: Each frame should look different in pose but still belong to the same person and fashion shoot.
- Overloading the prompt with adjectives: More descriptive words do not automatically mean better realism. Specific visual instructions are more useful.
- Allowing aggressive cropping: A fashion collage can lose its visual balance if one frame cuts through the head, shoulders, or body unexpectedly.
Making the Composition Work on a Phone
The prompt already accounts for available screen space, which is useful because a three-frame collage can become cramped when each section contains a full-body subject. The instruction to leave natural breathing room helps prevent the model from pushing the subject too close to the frame edges.
For social posts or mobile viewing, keep the subject clearly recognizable in each horizontal section. A beautiful background is secondary; if the person's face becomes tiny or heavily cropped, the three-frame concept loses much of its purpose.
When This Prompt Works Best
This setup is particularly suitable for fashion inspiration posts, social media portraits, profile-content concepts, editorial-style photo transformations, and aesthetic personal photography. It is less suitable when you need exact commercial product placement, precise typography, or a highly controlled graphic layout.
The prompt's strength is the combination of identity preservation, repeated character appearance, and natural fashion photography. Those three requirements are what make the editing workflow more demanding than a simple text-to-image fashion prompt.
Frequently Asked Questions
Can I use this prompt with a different person's reference photo?
Yes. The prompt is designed around a single reference image, so you can use another person's clear photo as the identity source. The quality and visibility of the reference face will still affect the consistency of the result.
Can I change the outfit without changing the face?
Yes. Outfit changes are already part of the prompt's design. Keep the face-preservation instructions unchanged and replace the clothing description with the style you want, while asking for the same outfit across all three frames.
Why does the prompt specify 3:4?
The 3:4 vertical format provides enough height for portrait fashion compositions while keeping the three horizontal frames inside one compact image. It also works naturally for portrait-oriented mobile content.
Final Thoughts
This prompt is strongest when the goal is not simply to generate an attractive fashion image, but to make a new editorial composition around an existing person without losing their recognizable identity. The combination of explicit facial preservation, consistent clothing and proportions, controlled framing, and natural outdoor photography gives the model a clear hierarchy of what should change and what should stay fixed.
For the most reliable result, start with a strong reference image and change one visual variable at a time when refining the output. That approach usually gives you more control than continually adding adjectives to an already detailed prompt.
