UNLIMITED Seedream 5.5 · Live on every paid plan

See how
Cover for the guide on turning a photo into a video with AI
Prompting

Official BytePlus partnerBytePlus, a ByteDance company

How to turn a photo into a video with AI

Turn a photo into a video with AI: add the picture, describe only the motion, pick the size and length. Clips run 4 to 30 seconds, from 17 credits.

KreatorFlow3 min read

Short answer

Open the video studio, add your photo, describe only the motion, and generate. The picture becomes the start of the clip, so you do not need to describe what is already in it. A 5-second clip costs 26 credits on Seedance 2.5 at 720p or 17 on Seedance 2.5 Flash.

Updated:

Photo to video in five steps

  1. 1

    Pick a clear photo

    A sharp photo with one clear subject works well. Avoid heavy filters and tiny faces in a crowd. Use a photo you own or have the right to use.

  2. 2

    Add it in the video studio

    Open the studio and add the photo. You can add up to 9 images in one clip if you want the model to see more than one angle.

  3. 3

    Describe the motion, not the picture

    The model can already see the photo. Write what moves: the camera, the wind, the water, the person's head. Keep it to two or three sentences.

  4. 4

    Match the shape

    Choose a size that fits the photo: 16:9, 4:3, 1:1, 3:4, 9:16, 21:9 are available. Pick the shape you plan to post in.

  5. 5

    Generate and compare

    Start with 5 seconds. If the motion is too much or too little, change one line and run it again.

Motion prompts for a photo

Add your photo, then paste one of these.

  • Portrait

    The camera moves in slowly. Her hair moves a little in the wind and she turns her head toward the camera and smiles.

  • Landscape

    Clouds move across the sky and the water ripples. The camera drifts slowly to the right. Soft sound of wind and waves.

  • Pet

    The dog looks around, then lifts its head and wags its tail. The camera stays still.

  • Product

    The camera circles the product slowly. Soft light moves across its surface. No other movement.

What works and what does not

Gentle motion is the safest: a slow camera move, wind, water, light, a small head turn. Large changes such as turning a person all the way around or adding many new objects are harder to hold together.

If a hand or a face looks wrong, make the motion smaller or shorten the clip. A 4 or 5 second clip has less room to drift.

Keeping the look of the photo

Say what must not change: the outfit, the colours, the background. One short line is enough, for example: keep the same clothes and the same background.

If you want the photo to talk or sing, use the lip sync tool instead. It is made for faces and speech.

Credits

A 5-second clip at 720p is 26 credits on Seedance 2.5 and 17 on Seedance 2.5 Flash. The studio shows the exact number before you press generate, and a failed generation returns its credits.

From code

The developer API takes a photo and a prompt too, if you want to run many photos at once.

Frequently asked questions

How do I turn a photo into a video with AI?

Open the video studio, add your photo, describe only the motion, and generate. The picture becomes the start of the clip, so you do not need to describe what is already in it. A 5-second clip costs 26 credits on Seedance 2.5 at 720p or 17 on Seedance 2.5 Flash.

Do I describe the photo in the prompt?

No. The model already sees it. Describe the motion: how the camera moves and what in the picture moves.

How long can the video be?

From 4 to 30 seconds on Seedance 2.5.

Can I use more than one photo?

Yes. The studio takes up to 9 reference images in one clip.

Will the video have sound?

Yes. Sound is part of the clip and adds no credits. Describe it in the prompt if you want a specific sound.

Can I make the photo talk?

Use the lip sync tool for speech or singing. It is built for faces and a voice track.

Animate your photo

Add the photo, describe the motion, and see the credits first.

More guides