Alley Selfie Reconstruction with Targeted Modifications
A JSON prompt to regenerate a photorealistic cobblestone-alley selfie from a reference image, preserving the same woman, hairstyle, makeup and lighting, while applying targeted edits: inner top to red, outer top to white, earring to red, removal of all neck jewelry and of the buildings at the top of the background.
Source previews
These example images come from the original source. DROPIDEA has not independently tested these examples; results depend on the model and settings.
Your prompt
Replace the red variables with your information. Copy the full prompt and use it with your preferred agent.
What you get
- One extremely photorealistic vertical selfie at roughly 4:5 to 9:16
- The same woman, face, slicked-back hair and soft glam makeup as the reference
- A white scoop-neck outer top with a red inner layer and a red earring
- A completely bare neckline with no necklaces or charms, and a clean upper background without building clusters
- Natural late-afternoon lighting with a genuine smartphone-photo look
Use cases
- Trying wardrobe and accessory color changes before committing to them
- Producing fashion or social-media portrait content with a European street feel
- Testing directive image editing that preserves a person's identity and features
- Creating reference imagery for showcasing hairstyles and makeup realistically
How to use
- Upload the original reference image (the alley selfie) with the request; the model relies on it to match the woman, her features and the background with high fidelity.
- Paste the template as-is without altering technical values such as aspect_ratio (approximately 4:5 to 9:16), focal_length_equivalent (24-28mm) or processing (HDR).
- If you want to adjust the colors, change only the color values at wardrobe.outer_top, wardrobe.inner_top and accessories.earrings.color, keeping the removal instructions intact.
- Generate the image in portrait format, then compare the result against the reference to verify the face, hairstyle and expression stay consistent.
- Retry with the same template if the result drifts; image generation is stochastic and an identical image cannot be guaranteed.
Related prompts
Four-Panel Streetwear Fashion Campaign
A prompt for creating a premium vertical four-panel streetwear campaign collage from an uploaded facial identity re...
Summer Beach Bar Photorealistic Portrait
A photorealistic image-generation prompt for a vertical summer portrait of a fictional adult Japanese woman in a st...
Fantasy Portrait with Chromatic Glitch
A prompt for a photorealistic fantasy fashion portrait featuring a woman with elven details, black and orange cloth...
Low-Angle Suburban Fashion Portrait
An ultra-photorealistic fashion editorial prompt featuring a low-angle outdoor portrait in a suburban setting with...