How to turn a photo into a video with AI

Turning a photo into a video takes three things: a good image, a short description of the movement, and a few minutes of rendering. Here is how to get a natural result.

Step 1: choose the picture

Pick a sharp, well-lit image with one clear subject and some space around it. Crop it to the shape you want, because the video will match it.

Step 2: describe the movement

Do not describe what is already in the picture. Describe what changes:

She looks up, smiles and brushes her hair back. A light breeze moves the curtain. Static camera.

Step 3: choose the length

Start with something short. Five to ten seconds is plenty for one small action.

Step 4: decide on sound

Describe ambient sound such as birdsong or a quiet room. Add a spoken line in double quotation marks only if the person in the photo has agreed to it.

Step 5: check the result

Watch the eyes, mouth and hands closely, since these are where problems appear. If something looks wrong, ask for less movement.

Why results look fake

• Too much movement for the amount of information in the photo.
• A blurry or very small source image.
• A subject cropped tightly at the edges.
• Busy group photos, where several people must all move at once.

Be respectful

Use only pictures you have the right to use, and get permission from the people in them. See our photo to video guide for more.