ChatGPT

Couple goals 😍

Written for ChatGPT. Results and reference-photo support may differ across models.

The complete prompt

Create a highly realistic two-panel vertical nighttime portrait collage using my uploaded photo as the ONLY identity reference for me. Use the reference image only for the poses, city setting, outfits, bouquet, lighting, composition and overall mood.

STRICT FACE & IDENTITY PRESERVATION: My uploaded photo controls my identity. Preserve my exact face β€” same face shape, proportions, forehead, eyes, eye spacing, eyebrows, nose, lips, cheeks, cheekbones, jawline, chin, skin tone, natural asymmetry and recognizable facial features. Do not copy the reference person's face. No face swap, face morphing, generic AI face, celebrity resemblance, beautification, reshaping or facial-feature alteration.

Overall composition

Create a vertical 9:16 collage with two horizontal photographs stacked vertically, separated by a thin clean dividing line.

Both panels should look like authentic candid nighttime photographs taken in the same city location, with consistent lighting and environment.

Setting

Place me on a clean, wide city road at night, surrounded by landscaped greenery, neatly trimmed round bushes, trees and modern high-rise buildings with illuminated windows.

Include realistic urban details such as:

bright street lamps

traffic signals

road markings

distant illuminated signs

modern buildings

subtle nighttime bokeh


Keep the road realistic and naturally illuminated. The background should be slightly out of focus while I remain sharply detailed.

TOP PANEL β€” Bouquet portrait

Show me standing on the roadside holding a large bouquet of soft pink hydrangea flowers wrapped in transparent cellophane.

The bouquet should contain several full clusters of delicate pink hydrangeas with realistic green leaves and stems.

I wear an oversized black long-sleeved sweatshirt/top, with a relaxed casual fit.

Hold the bouquet naturally against my torso with both hands.

Turn my body slightly sideways while looking toward the side rather than directly into the camera, creating a calm candid expression.

Keep my hair natural, long and dark, falling softly around my shoulders.

BOTTOM PANEL β€” Playful pose

Show me at the same nighttime location, wearing the same fitted/simple black long-sleeved top.

Hold the same style of large pink hydrangea bouquet wrapped in clear cellophane in front of my body.

With my other hand, raise my sleeve-covered hand gently across my forehead/eyes in a playful, slightly shy pose. My eyes are softly closed and I have a subtle natural smile.

Keep the pose relaxed and candid, as if photographed spontaneously under the streetlights.

Add a small delicate necklace if appropriate, keeping the jewelry minimal.

Hair

Keep my natural hair identity and color from my uploaded photo. Hair should look soft, naturally styled and slightly tousled by the outdoor breeze, with realistic individual strands catching the streetlight.

Lighting

Realistic nighttime city lighting:

warm white/yellow streetlights

subtle cool ambient blue tones from the night sky

realistic highlights on the bouquet's transparent wrapping

soft reflections on the road

natural shadows

gentle rim light around the hair and shoulders


Create a cinematic but believable nighttime atmosphere without making it look artificially edited.

Photography style

Ultra-realistic DSLR/social-media photography, authentic nighttime candid photo, natural skin pores and texture, realistic hair strands, accurate hands and fingers, detailed hydrangea petals, realistic transparent plastic wrapping, natural fabric texture, shallow depth of field, subtle bokeh, realistic streetlight bloom, high-resolution detail, soft cinematic contrast and very subtle film grain.

It should look like a genuine photograph taken on a smartphone/DSLR at night, not an AI-generated image.

Identity priority

MY UPLOADED PHOTO = ONLY SOURCE FOR MY FACE AND IDENTITY.
REFERENCE IMAGE = SOURCE ONLY FOR POSES, CLOTHING, bouquet, environment, lighting and composition.

The same person must appear consistently in both panels.

Negative prompt: different face, reference person's face, different person, face swap, face morphing, generic AI woman, celebrity resemblance, altered facial structure, beautified face, plastic skin, excessive smoothing, unrealistic eyes, distorted hands, extra fingers, malformed bouquet, artificial flowers, warped body, fake city, CGI, cartoon, 3D render, excessive HDR, blurry face, inconsistent identity, watermark, text.

Copy the complete prompt and use your own reference photos where requested.

Open ChatGPT β†—

Example image provided with this prompt. This prompt has not been tested by our site.

Photo studio
AI-generated example

A little inspiration for your own photo.

1Add your photos
Person A

Keep the face clearly visible

Person B

Keep the face clearly visible

JPG, PNG or WebP Β· Max. 10 MB each

Your photos are sent only when you choose Generate. Use photos you have permission to use.

2Make it yours
Change the details
Edit your prompt
Open Gemini

Getting ready…

Sign in to your account

Continue with your Google account.

You can browse and copy prompts without signing in.

How we use your information: Privacy policy