Preserve each person's recognizable identity from the reference: eye shape, nose, mouth, face shape, skin tone, approximate age, distinctive features, and natural skin texture. Change only the hairstyle, clothing, pose, lighting, and background. For group photos, map each face to the correct person, keep the exact number of people, and prevent identity blending or face swapping. Do not make absolute promises about an identical face. Treat each reference face as a separate identity and preserve the stated four-person order before applying the living-room snapshot scene. Keep the approved clothing, pose, and background unchanged.
1986 · family
1980s Family Living Room Snapshot
A candid family-of-four living-room photograph with period furniture, on-camera flash, and natural documentary energy.
The preview image shows the intended direction. It is an internal visual baseline, not a claim that ChatGPT or Gemini produced this exact frame.


Full prompt
Create a candid 1986 family photograph using the uploaded reference images. Show exactly four family members together in one living room, preserving each person's identity and their left-to-right order from the references: adult 1 on the left, child 1 seated in front, child 2 seated beside them, adult 2 on the right. Use believable 1980s living-room details such as a patterned sofa, wood panel accents, a CRT television, and a modest holiday decoration, with no modern electronics or logos. Light the scene with direct on-camera flash and warm household lamps, natural imperfect smiles, relaxed overlapping poses, and a 35mm consumer color-film snapshot look with fine grain and mild red-eye risk kept under control. Use a horizontal group composition with enough headroom and all hands visible.
Make it yours
Create a candid 1986 family photograph using the uploaded reference images. Show exactly four family members together in one living room, preserving each person's identity and their left-to-right order from the references: adult 1 on the left, child 1 seated in front, child 2 seated beside them, adult 2 on the right. Use believable 1980s living-room details such as a patterned sofa, wood panel accents, a CRT television, and a modest holiday decoration, with no modern electronics or logos. Light the scene with direct on-camera flash and warm household lamps, natural imperfect smiles, relaxed overlapping poses, and a 35mm consumer color-film snapshot look with fine grain and mild red-eye risk kept under control. Use a horizontal group composition with enough headroom and all hands visible. Subject: family; gender expression: mixed; aspect ratio: 4:5; composition: group; model: chatgpt. Show exactly the requested family arrangement and keep each person mapped to their reference position. Preserve each person's recognizable identity from the reference: eye shape, nose, mouth, face shape, skin tone, approximate age, distinctive features, and natural skin texture. Change only the hairstyle, clothing, pose, lighting, and background. For group photos, map each face to the correct person, keep the exact number of people, and prevent identity blending or face swapping. Do not make absolute promises about an identical face.
Fix a result
Needs reference: Original + result
Preserve each person's recognizable identity from the reference: eye shape, nose, mouth, face shape, skin tone, approximate age, distinctive features, and natural skin texture. Change only the hairstyle, clothing, pose, lighting, and background. For group photos, map each face to the correct person, keep the exact number of people, and prevent identity blending or face swapping. Do not make absolute promises about an identical face. Treat each reference face as a separate identity and preserve the stated four-person order before applying the living-room snapshot scene. Keep the approved clothing, pose, and background unchanged.
Identity lock
Preserve each person's recognizable identity from the reference: eye shape, nose, mouth, face shape, skin tone, approximate age, distinctive features, and natural skin texture. Change only the hairstyle, clothing, pose, lighting, and background. For group photos, map each face to the correct person, keep the exact number of people, and prevent identity blending or face swapping. Do not make absolute promises about an identical face.
Why these details matter
This prompt combines a specific year, a photographic situation, a light source, wardrobe, composition, and film behavior. Those details give an image model observable evidence instead of asking for a vague “retro” effect. The identity sentence stays separate from the styling so a reference can remain recognizable while the setting changes.
The scene also avoids readable logos and contemporary devices. That keeps the result adaptable and reduces accidental brand or text artifacts. If your model introduces a modern object, remove it with a focused correction before changing the rest of the composition.
Correction prompts
Use the reference image as a strict map: preserve each person's face and place each person in the same left-to-right order. Do not blend identities or swap facial features.
Increase the authentic 1980s photographic language: period-accurate film color, on-camera flash behavior, optical softness, era-appropriate styling and props. Avoid a generic neon cyberpunk look.
Remove contemporary objects, logos, devices, cars, and architecture. Replace them with details plausible for the specified 1980s setting while preserving the subject's identity.
How to use this prompt
Choose a clear reference image, paste the complete prompt into your image model, and start with one of the listed aspect ratios and compositions. Check the year, number of people, identity, light direction, and crop before making a style judgment. If a result drifts, use one correction prompt at a time so you can see which change helped.
Model notes
ChatGPT and Gemini may interpret the same wording differently. Keep the identity lock and the scene sentence intact, then adjust only the model-specific image controls. The preview here is a visual baseline for the idea; run your own test before treating a combination as reliable.
FAQ
Which models work?
These prompts are formatted for ChatGPT and Gemini image generation, but the photographic instructions can be adapted to other models.
Will the face be identical?
Identity guidance improves consistency but no image model can guarantee an identical face across every generation.
Can I use the result commercially?
Check the model provider’s terms and make sure you have permission for any reference image, person, logo, or other source material.