Canonical page: https://nureta.ai/en/guide/image-to-video-prompts

# Image to Video Prompts: How to Make Your Photo Move, With 15 Examples

By Nureta Team · Updated November 21, 2026

When you turn a picture into a video, the picture already shows who is there and where. Your prompt has a different job: it says what moves, how the camera moves, and how the shot ends. On Nureta, a picture becomes a video in two ways: as a face photo of a person in your cast, or as the first or last frame of a shot. Each way needs a slightly different prompt. This guide shows what to write for each, with 15 examples that you can copy.

Nureta is the No.1 unfiltered NSFW AI video platform, trusted by 100,000+ users. Text to video and image to video, up to 3 minutes.

## Two ways your picture is used, and what to write

Face photos: under Who's in it on Create, each person card takes a photo of a consenting adult, for up to 3 people. The photo sets who the person is, and the same faces stay in every scene. You write a normal story: the place, the action and the camera.

First and last frames: for a video longer than 15 seconds, set Mode to Edit on the confirm step. In the storyboard, each shot of 4 to 15 seconds has a short text that says what happens in it, and it can take your picture as its first frame, its last frame or both. A frame is used as the exact image, so the text describes the motion, not the picture.

You choose one way per video: frames cannot be combined with face photos, saved Library characters, reference images or backgrounds.

| Way | Where you add the picture | What your prompt does |
| --- | --- | --- |
| Face photo | A person card under Who's in it | Tells the story. The photo sets who it is |
| First frame | A shot in Edit mode (videos over 15 s) | Says how the picture starts to move |
| Last frame | A shot in Edit mode | Says what happens before the shot lands on the picture |
| First and last frame | The same shot in Edit mode | Says how to get from one picture to the other |

## The 4 things a motion prompt should name

Describe what changes, not what the picture already shows. The frame already has the hair, the dress and the room, so spend your words on the movement. One clear action for every 4 or 5 seconds usually works better than five actions in one shot.

Name these four things:

| Part | Example words |
| --- | --- |
| The person moves | turns her head, smiles slowly, stands up, walks toward the camera |
| The camera moves | slow push-in, the camera holds still, pans left, handheld, close-up |
| The world moves | curtains in the breeze, rain on the window, candlelight flickers, waves |
| Pace and ending | slow and dreamy, calm, urgent, ends on a close-up of her eyes |

## First frame: start from your picture (examples 1 to 5)

Upload your picture as the First frame of a shot. The shot opens on it and moves from there.

1. She turns her head toward the camera, smiles slowly and brushes her hair back. Slow push-in. Soft, warm light.

2. He stands up from the chair and walks toward the camera, loosening his tie. The camera holds still. Calm and confident.

3. The curtains move in the breeze. She stretches on the bed and rolls onto her side, looking at the camera. A gentle, dreamy pace.

4. Rain runs down the window behind her. She takes a sip of wine and steps closer. Slow push-in, then a close-up on her eyes.

5. She laughs, takes his hand and pulls him out of the frame toward the bedroom. A handheld camera follows them.

## Last frame: end on your picture (examples 6 to 8)

Upload your picture as the Last frame. The shot moves toward it and ends on it, so write what happens before that moment.

6. She walks in through the door, drops her coat and ends sitting on the edge of the bed, looking at the camera.

7. The camera starts wide on the room and slowly moves in until it ends on a close-up of his face.

8. They dance slowly across the room and end in the embrace shown in the picture.

## First and last frame together (examples 9 and 10)

Upload one picture as the first frame and another as the last frame of the same shot. The text says how to get from one to the other, so keep the two pictures in the same place and with the same people.

9. She gets up from the sofa, walks to the window and turns to face the camera. Smooth, slow movement.

10. They move from the doorway to the bed, kissing, in one slow, continuous shot.

## Face photos: put your cast in a scene (examples 11 to 13)

With face photos, the photo sets who the person is and the story sets everything else. Use the same names in the story as on the person cards.

11. Lena and Jon at a candlelit dinner on a terrace. Lena leans across the table and kisses him. Close-up, warm light.

12. POV of Jon. Lena lies back on white sheets, looks up at the camera and says {Come here.} Soft morning light.

13. Scene 1 (Hotel bar): Lena and Jon flirt over cocktails. Scene 2 (Elevator): they kiss as the doors close. Scene 3 (Room): slow and sensual, in lamp light.

For more story prompts, read [24 NSFW AI video prompt examples](https://nureta.ai/en/guide/nsfw-ai-video-prompt-examples).

## Long videos: several shots, several pictures (examples 14 and 15)

In Edit mode, every shot can take its own frames, in a video of up to about 3 minutes. Auto chaining is off by default. Turn it on for smoother joins: each clip then starts from the last frame of the clip before it in the same place (a new place is still a cut), but the video renders more slowly.

14. Shot 1, first frame of her at the door: "She opens the door and smiles." Shot 2: "They walk into the room, kissing." Shot 3, last frame of the two of them on the bed: "They fall onto the bed, laughing."

15. Shot 1, first frame of her on the beach: "She walks along the shore toward the camera." Shot 2: "She sits on the sand at sunset as the waves roll in." Shot 3: "Close-up. She looks at the camera and smiles."

## When the video does not look right

Most problems have a simple fix in the prompt or in the storyboard:

| Problem | What to change |
| --- | --- |
| Nothing moves, or very little | Name one clear action and one camera move in the shot text |
| Too much happens, it looks rushed | Keep one action per shot, or make the shot longer (up to 15 s) and another one shorter |
| A face changes on a shot with a frame | A shot with a frame ignores its face references. Use a frame that shows the face clearly |
| The background is ignored | A shot with a frame ignores its background image. Put the place in the frame itself |
| The joins between shots jump | Turn on Auto chaining (the render is slower) |
| A face photo does not look like the person | Use a clear, well-lit, front-facing photo of one person |
| One clip came out wrong | Fix a part remakes that clip. Each paid video includes 1 to 3 free clip fixes within 3 days |

## Which pictures you may use

Use only pictures you have the right to use, of consenting adults: yourself, or a partner who explicitly agreed. An automatic age check runs on every photo and frame, and photos of minors are rejected. Celebrities and public figures are not allowed, and photos that look like one are rejected automatically. A photo of a copyrighted character is refused too.

Upload a JPG, PNG, WebP or HEIC file of up to 20 MB. Saved Library items cannot be used as frames yet, so upload the file itself. The free first video is a 5-second clip, so it has no Edit mode and no frames, but it can use face photos. For all the rules, read [adult AI video from your own photos: the consent rules](https://nureta.ai/en/guide/adult-ai-video-from-your-own-photos-consent-rules).

## Frequently asked questions

### What should I write in an image-to-video prompt?

Write what moves: the person's action, the camera move, what moves around them, and how the shot ends. For example: "She turns her head toward the camera, smiles slowly and brushes her hair back. Slow push-in." One clear action for every 4 or 5 seconds usually works well.

### Do I need to describe what is in my photo?

Not much. A first or last frame is used as the exact image, so its people, clothes and room are already there. Spend your words on the motion and the camera. With face photos, the photo sets the face, and your story sets the place and the action.

### Can I make a long video from one photo?

Yes. A face photo keeps the same face in every scene of a video of up to about 3 minutes. In Edit mode, you can also give every shot its own first or last frame.

### Why can't I add a first frame?

Frames need Edit mode, which is for videos longer than 15 seconds, so the free 5-second clip has no frames. Frames also cannot be combined with other pictures on the Create page: to use frames, remove the face photos, saved Library characters, reference images and backgrounds.

### Does image to video cost more than text to video?

No. The price depends on the length and the resolution: from $1.29 per 5 seconds at 480p. A creator's catalog scene can add a scene fee, and the confirm step shows it. Edit mode and frames cost nothing extra, and there is no subscription.
