Iteratively Edit a Coffee-Shop Tabletop Photo

Make a series of focused edits to a coffee-shop tabletop photo, changing one thing per round while using the previous result as the next starting point. The prompt includes a specific sequence of edits and a customizable current edit

LIQ AI Republished on Nov 16 0 Bookmarks 0 Copies
Bookmark 0

Example outputs

AI-generated video of a coffee-shop tabletop photo is edited one change at a time, from a coffee-to-orange-juice swap to a croissant, watch, sunglasses, and handwritten note.

Download video result 1

Prompt template

Usage instructions
Format
Text Parameterized
Inputs
Original_photo (image) · Required Current_image (image) · Required current_edit (text) · Optional · Default: replace the coffee with orange juice
Outputs
Edited_photo (image)
Edit the provided coffee-shop tabletop photo. Make only the change specified for this round: ${current-edit:replace the coffee with orange juice}. Keep the rest of the image as close to the current input as possible.

Use the previous round’s result as the current image for each subsequent edit, not the original photo. Apply these specified changes one per round, in order:
1. Replace the coffee with orange juice
2. Add a croissant
3. Replace the watch with a steel watch
4. Replace the receipt with sunglasses
5. Write “Bon dia, València” on the napkin

After each edit, compare the result with the original photo for changes in lighting, color, and details that were not meant to change

Replace the ${…} placeholders with your values before using this prompt. See parameters and defaults below

Parameters

Replace each placeholder with your own value, or use its default when one is provided

${current-edit:replace the coffee with orange juice} current-edit Default: replace the coffee with orange juice

What you'll create

This prompt guides a sequence of edits to a coffee-shop tabletop photograph, including replacing the coffee with orange juice, adding a croissant, and changing other objects. Each round applies one change, with the previous result serving as the next starting image

Usage instructions

Provide the original photo and the image for the current round; for the first round, use the original photo for both. Change the current edit input to the next listed change after each round, and compare each result with the original for shifts in color, lighting, or untouched details

Comments

Tried this prompt? Share what you made, compare results, or ask a question

Up to 4 results. Images up to 2 MB; MP4, WebM, or MOV videos up to 4 MB each