What is the difference between Motion control and Full body?

Motion control takes the person, clothing, and scene from the reference photo and the actions and expressions from the video. Full body brings the reference person’s look into the source video’s scene. Choose based on which background you want to use.

Does Motion control use the photo’s background?

Yes. The photo supplies the setting, but the generated movement may change framing and fine details. It does not guarantee a pixel-for-pixel match to the photo.

What makes a good motion video and reference photo?

Use a video with one clearly visible main subject and visible movements. Choose a sharp reference photo with an unobstructed face. For full-body movement, show the complete person and outfit, keep hands clear, and leave space around the subject.

Which aspect ratio will the result use?

The result follows the reference photo within the supported aspect ratios. The portrait photo in this tutorial produces a vertical result. Framing within that format can change during generation.

Is the original audio kept?

Yes. The generated result retains the motion video’s original audio. Some comparisons in the tutorial are muted so the narration is audible; that does not mean the downloaded result is silent.

How much does Motion control cost?

At the time of this tutorial, 720p costs 4 credits per second and 1080p costs 6 credits per second, with partial seconds rounded up. An 8-second clip at 1080p costs 48 credits. Check the current estimate before every submission; another generation is charged separately.

Can I adjust an existing result?

Use Re-generate to bring the inputs and settings back into the form. Make changes, check the new estimate, and submit another paid generation.

How to Animate a Photo with Motion Control

Make the person in a photo move using a video performance. This tutorial uses one 8-second motion clip and a courtyard portrait to show which details come from each input.

Use photos and videos you own or are authorized to edit, with permission to use the person’s likeness.

Open Motion Control

Open Face Swap, choose the Video tab, and select Motion control. Your reference photo supplies the person and setting; a motion video supplies the actions and expressions.

1Face SwapOpen the tool

2VideoChoose the video tab

3Motion controlSelect the mode

Motion control is a mode inside Face Swap’s Video tab.

Add a Motion Video

Upload or drag in a clip from your device, choose a video from Assets, or hover over a Trending Template to preview it. This example uses an official template.

Start with a short clip featuring one clearly visible main subject and the movements you want to transfer. The current workflow accepts videos from 4 to 180 seconds, up to 200 MB. MP4 is recommended.

The example motion video, showing one clear main subject — The example uses an 8-second template. Its movement and original audio carry into the result.

Choose Your Reference Photo

Under Reference Image, choose Upload, Assets, or Avatar Library. Pick a sharp photo with an unobstructed face. For full-body movement, show the complete person and outfit, keep the hands clear, and leave room around the subject.

Here, the reference is a person in a blue suit standing in a sunny courtyard. The courtyard will supply the result’s setting.

Full reference photo with a clear face, complete blue outfit, and courtyard background — Keep space around the subject so the movement has room.

Understand What Comes from Each Input

The photo provides the person, clothing, and background. The video provides the actions and expressions. This differs from Full body, which uses the source video’s scene.

Illustration of the source performance in an indoor room — Illustrative comparison: motion from the video, with the person and setting from the photo. Images are reframed for comparison.

Illustration of the reference person in a blue suit in a sunny courtyard — Illustrative comparison: motion from the video, with the person and setting from the photo. Images are reframed for comparison.

The output aspect ratio follows the reference photo within the supported formats. This portrait photo produces a vertical result. The performance can change the person’s size and framing; it does not preserve every pixel of the photo.

To keep the original video scene instead, see the Face only and Full body tutorial.

Set the Resolution and Check the Cost

Choose 720p or 1080p. At the time of this tutorial, 720p costs 4 credits per second and 1080p costs 6 credits per second, with partial seconds rounded up.

Check the estimate beside the generate arrow before submitting. This 8-second clip at 1080p costs 8 × 6 = 48 credits for one generation.

720p and 1080p options, 48-credit estimate, and purple generate arrow — One 8-second generation at 1080p. Always check the current estimate in the app.

Generate, Review, and Download

Click the generate arrow. When the result is ready, open its card in Videos below the form. Compare the face, hands, movement, outfit, and scene with your inputs throughout the clip.

The original video audio is preserved. This workflow does not clone or replace the speaker’s voice.

Actual Motion control result with the blue outfit and courtyard setting — The result keeps the original video audio.

Download icon below the result title in Video Viewer — The result keeps the original video audio.

Adjust and Generate Again

Click Re-generate to bring your inputs and settings back into the form. Change what you need, check the updated estimate, and submit a new generation. Submitting again is a new paid generation, not a free retry.

1Re-generateReturn to the form

2Edit your inputsAdjust the photo, clip, or settings

3Check and submitReview the new credit estimate

Returning to the form lets you edit. Submitting creates a new paid generation.

Try It with Your Own Photo

Choose a photo with the person and setting you want, then pair it with a clear motion video. Use the Try Motion Control button above to open the right mode.

For dance-focused examples, explore the AI Dance tool page.