AI-generated example image for demonstration purposes. AI-edited example image for demonstration purposes.
Photo Editing Prompts
Transform your website's fine art catalog with this breathtaking, museum-quality conceptual mosaic prompt. Harnessing a brilliant optical illusion, this template uses thousands of highly detailed miniature people as living pixels to meticulously craft a perfectly recognizable human face. The clever transition from a tight crowd configuration to scattered individual figures at the edges creates an incredible surreal depth. It provides digital artists and creative directors with a masterful visual narrative exploring humanity, connectivity, and collective identity.
Aspect Ratio
Select before copying
Recommended: 4:5 — 4:5 vertical framing comfortably accommodates two subjects while maintaining intimate composition and enough headroom for both faces.
Rate this prompt
Free prompt — one-click copy, no login required
Lighting
Set on a clean white minimalist background with soft natural lighting and long cinematic shadows cast by each tiny person from a top-down ae
Color
Extraordinary conceptual fine art human crowd mosaic portrait of a young man, preserving 100% exact facial structure, hairstyle, beard, skin
Step 01
Use the copy button above to grab the complete prompt with your selected aspect ratio.
Step 02
Use a generator that supports the input and reference-image features you need.
Step 03
Create variations, pick the best result, then download and use it in your content.
Issues you may encounter and how to fix them
Why it happens
Dual-subject identity preservation is significantly harder than single-subject — most models prioritize the face they detect first and treat the second as optional.
Fix
Specify separate identity-preservation instructions for each person (e.g., "person on the left must match reference face A, person on the right must match reference face B") and use a higher-quality reference photo for each.
Why it happens
Full-body portraits require the model to correctly render hands, feet, and body proportions simultaneously — a known challenge for AI image generators.
Fix
Specify "natural relaxed pose, correct body proportions, anatomically coherent limbs and hands" and keep the pose simple to minimize proportion errors.
Why it happens
Some AI models prioritize artistic interpretation over photorealistic rendering, especially when the prompt contains both stylistic and technical instructions.
Fix
Move all camera and quality specifications to the beginning of the prompt and add "ultra-sharp focus on the subject, fine detail rendering, professional photography quality" to reinforce the desired output.
Adapt these elements to fit your vision
This prompt works with popular AI image generators.
ChatGPT
Prompt helper
Midjourney
Image model
Leonardo AI
AI creator
Gemini
Google AI
Stable Diffusion
Open model
Links to official tool websites. Not a claim that this prompt was tested in every tool or model version.
Quick answers before you copy this prompt
Yes, this prompt is designed for reference-image workflows. Upload a clear, well-lit photo of the subject. If you remove the identity-preservation instructions, the model will generate a new face instead.
Use separate reference photos for each person and specify which reference corresponds to which person in the prompt. For example: "the woman must match reference photo A, the man must match reference photo B." Quality reference photos with clear facial features work best.
Use 4:5 for Instagram feed posts (best vertical coverage), 1:1 for profile pictures and square grids, and 9:16 for Stories and Reels. For landscape platforms like Twitter/X, try 16:9 but note the full-body composition may need adjustment.
Add "natural skin texture with visible pores, subtle imperfections, realistic skin tone variation" to the prompt. Avoid words like "flawless" or "perfect skin" which push models toward artificial smoothness.
