How it works
From two photos to one picture.
What to upload, what the AI does with it, and how to get a result worth keeping.
1. Choose the photo to complete
This is the scene: a group shot, a holiday photo, a family portrait. It works best when there is free space next to the people already in it, and when faces are not tiny. The picture keeps the shape of this photo: wide, tall or square.
2. Choose a photo of the person to add
Use a photo that shows only that person, sharp and evenly lit, looking toward the camera. A photo from the waist up or full length gives the AI more to work with than a tight selfie. Light from a similar direction to the scene helps.
3. Say where they should stand
Choose left, middle or right, or leave it to the AI. A short note can describe a pose, such as standing behind the sofa, or holding the baby.
4. Check the free preview
The preview is smaller and watermarked. Compare each face with the original photos. The AI redraws the whole image, so people who were already in the photo can change too.
5. Download the original
Pay US$1 through Stripe and download the full-size picture without the watermark. If you gave an email address, we send you the download link. The picture is deleted 72 hours after it was made.
Tips for a better result
Use photos taken in similar light: both outdoors, or both indoors. Avoid sunglasses, hats that shade the face and heavy filters. If the person looks too large or too small, add a note such as "standing at the back". Every preview is different, so if one is close but not right, create another.
Ready to try it?
The preview is free.
Open the studio ↗