AI Editor

Add or remove objects in any photo with a few words

Your users name the object to take out or put in. AI edits that part of the photo, and the design around it stays as it was.

  • Removals and additions from one instruction
  • Before and After on every edit
  • Works on customer uploads in print templates
Get in Touch
Available for
Web
Mobile
Desktop

From a cluttered shot to a clean one

Retouching takes skills most customers don’t have. Masking a lamppost out of a holiday photo is a job for a pro. With object edits, it’s a job for anyone who can type.

Name what should go or come

Your users write which object to remove or add. It works on any photo in the design, from product shots to their own uploads.

Compare it with the original

The photo is edited where it sits, with Before and After side by side until the change is kept or discarded.

Fix what the first pass missed

The result is still a regular photo. Users send a second instruction for anything left over, then crop or filter it as usual.

Trusted by 100+
industry leaders

Digitas
Omneky
The Print Bar
Brandwatch
Planoly
HP
Shopify
Reuters
Hootsuite
Semrush
Shutterfly
one.com
Sprout Social
Constant Contact
Spread Group

Removing and adding, in plain words

Remove

Take out what distracts

Users name the object and where it is, like “the person in the red jacket on the left”. The model finds it and rebuilds the scene behind it.

Add

Put in what was missing

Users say what to add and where it belongs, like “a small potted plant on the desk next to the laptop”. The model places it in the scene.

Set up object edits: model and template

Step 01

Turn on image editing

Object edits are part of image to image, so adding the AI image generation plugin turns them on for the web, iOS, and Android.

The same photo selected in the editor on Android, iOS, and the web
Step 02

Pick a model that changes one thing

An object edit only works if everything outside the object stays as it was. The Best GenAI Models for Web-to-Print compares how each editing model handles that kind of request.

Model picker listing AI models next to a design canvas
Step 03

Leave the photo open in print templates

Customers can only fix an upload if the template lets that photo be replaced. Keep the placeholder editable and lock the rest.

Smart print templates with placeholders and lockable layouts

What teams add and remove in their photos

Customer photos ready for print

A shopper takes a stranger out of the background before the photo goes on a canvas print.

Reusing older images

A designer takes the date stamp off an archive photo and reuses the photo in a new design.

Props for product shots

A marketer adds the coffee cup that was missing on set, and the shot works for a lifestyle banner.

Seasonal versions

A retailer adds snow on the windowsill and string lights, and the everyday storefront photo becomes a winter image.

More AI features

Image to Image

Change a photo that is already in the design with a written instruction.

Style Transfer

Redraw any image as a painting, a cartoon, or a realistic photo.

Text to Image

Generate a new picture from a prompt, right on the canvas.

AI Sticker Generation

Create a sticker in a chosen icon style when the library has none that fits.

Get in touch

Take something out of a photo in the AI Editor demo

Select a photo, describe one object to remove or add, and check the result. Then talk to us about adding object edits to your product.

Frequently Asked Questions

AI object removal takes an unwanted object out of a photo, and the same instruction-based edit adds something that's missing. The user names the object in words, and an editing model repaints that part of the photo. In CE.SDK, it runs as an image to image edit on the photos in a design.

Inpainting usually asks the user to paint a mask over the object first. Here the user describes the change in words, and the editing model locates the object itself.

Yes. An upload becomes a regular photo in the design, so the same Edit Image action works on it.