PhotoImage outputGuided
Multi-Person Group Photo Cleanup
Clean a group photo (lighting, clutter, mild distractions) without changing anyone's recognizable face or inventing people. Why it matters: a clear picture beats a confusing one.
Scanned · no issuesReady to use
Prompt
Clean my group photo (lighting, clutter, mild distractions) without changing anyone's recognizable face or inventing people.
Context handling:
- First use relevant information already available from the conversation, attachments, or connected sources.
- Do not ask me to repeat details you already have.
- If a required detail is missing, ask one short follow-up.
- State important assumptions. Never invent identity details.
Group face lock (non-negotiable):
- Preserve every recognizable person's exact face, skin tone, hair, age cues, and expression.
- Do not beautify, face-swap, average, or replace any face.
- Do not invent new people. Do not remove people by default.
- Hard rule: never replace faces or add/remove people unless I explicitly ask to remove a photobomber AND clearly point them out (see remove_photobomber). Keep natural pores and realistic skin—no plastic skin.
Default look unless I override variables:
- Cleanup notes: {{cleanup_notes}}
- Remove photobomber: {{remove_photobomber}}
- If cleanup_notes is empty, gently even lighting, reduce glare, and clean mild background clutter/distractions.
- If remove_photobomber is empty, do not remove anyone.
Hard bans:
- No readable text, logos, watermarks, badges, brand marks, or captions in the image (unless tiny unreadable UI chrome is genuinely required by the product scene).
- No plastic/over-smoothed skin, over-whitened teeth, mismatched lighting, or surreal effects.
- No identity changes, no fabricated attendees.
Produce:
One high-resolution cleaned group photo with all identities preserved and no caption or border.Create your final prompt
Add your context and generate
Fill in the required variables below. Optional variables can be left blank and will remain as placeholders.
What to clean (clutter, lighting, distractions)
Only if user explicitly points out a photobomber to remove
Published example
Sample image output

The same group with cleaner lighting and less clutter, every recognizable face preserved, no invented people, no face swaps.
Example context
Example input
Upload a group event photo. Optionally set cleanup_notes to even lighting and remove edge clutter. Only set remove_photobomber if you explicitly identify someone to remove.
Old photo restoration prompts
More in this series
Keep exploring
Discussion
Comments (0)
Sign in to leave a comment.
Loading comments…