Genjutsu · Image to video generator

Image to Video AI

Most AI video tools start with a blank prompt and ask you to imagine everything from scratch. This one can start with a photo you already have. Upload an image, describe how it should move, and download a short clip for a post, a product listing, or someone you care about. No editing timeline or special software required.

How to turn a photo into a video

Three steps, no software to install.

  1. 1

    Upload your photo

    Choose a clear JPG, PNG, or WebP photo, up to 12 MB on supported image-to-video inputs. Some models accept fewer formats; the uploader checks your chosen model. A sharp, well-lit original gives the model more detail to work with.

  2. 2

    Describe the motion

    Tell the model what should move or how the camera should travel. Try “slow zoom in on the face”, “camera pans left across the room”, or “hair moving in a gentle breeze”.

  3. 3

    Generate and download

    Choose the available duration and resolution for your model, review the credit estimate, and generate. When the video is ready, download the MP4 or try another variation.

Photo to video, image to video, picture to video — the same job

People look for this feature under different names, but the job is the same: give a still image movement. Whether you call it photo to video, image to video, or picture to video, you start from a frame you have already chosen instead of generating a whole scene from scratch.

What changes is the motion you ask for. A camera can push closer, drift across a room, or rise slowly above the subject. A person might turn their head, blink, or feel wind moving through their hair. The model interprets those instructions and creates a new short clip. The starting picture helps guide the result, although fine details are not guaranteed to remain perfectly unchanged.

What people use it for

You do not have to film something new every time you want movement. Start with an image you already like and think about what a few seconds of motion would add.

Social posts

A still product shot or portrait can feel static in a feed. A gentle zoom or pan gives it movement for Reels, Shorts, and other posts without having to record new footage.

Product listings

Turn a clean product photo into a short showcase. Subtle camera motion can help a listing stand out, though generated views are not a substitute for accurate product photos.

Old family photos

Give a scanned family picture a gentle push-in or a small atmospheric movement. Keep the motion restrained if preserving the original faces matters more than making the scene dramatic.

Real estate

A slow pan across an exterior or interior image can give a listing page a sense of movement without a dedicated video shoot. Review the result for invented structural details.

Covers and thumbnails

Make a moving background for a channel intro, book announcement, or promotional cover. Starting from your finished image helps preserve the overall composition.

Ads and landing pages

Try a few camera directions on the same creative, then keep the one that works best for your audience. Each generated variation uses credits, so start with a simple motion idea.

Image to video, or text to video?

They solve different problems. Image to video gives the model a picture to work from. It is useful when you already have a portrait, product, illustration, or scene and want to keep its overall look while adding motion. Composition, lighting, and the subject all have a starting point.

Text to video starts from a description instead. The model has more freedom to invent framing, lighting, and subject details. If you have no image yet and want to explore a scene from scratch, start with our text to video generator. The two workflows can complement each other, but an existing image is usually the better starting point when a specific composition matters.

Five ways to get a better result

  1. 1

    Start with a sharp, evenly lit photo.

    Motion can magnify blur, grain, and visual artifacts that were barely noticeable in the original.

  2. 2

    Leave room around the subject.

    A pan or zoom can crop the edges. Give faces, products, and other important details some breathing room.

  3. 3

    Describe the camera, not just the mood.

    “Slow dolly forward” or “camera pans left” gives the model a clearer instruction than “make it cinematic”.

  4. 4

    Keep faces and hands under control.

    Fast motion and extreme close-ups are harder to keep consistent. Try a wider view and gentler movement.

  5. 5

    Try more than one generation.

    The same image and prompt can produce different results. Compare a few runs when the shot really matters.

Is photo to video free?

You can open the generator and explore the controls without paying up front. Generating a video consumes credits, and the required amount changes with the model, duration, and resolution. The estimated credit cost is shown in the editor before you submit a job.

If your account has trial or promotional credits, you can use those toward an eligible generation. Otherwise, you can purchase credits as needed. There is no blanket promise of a free video allowance, fixed render time, or watermark-free output for every model. Check available credit packs before you generate.

Frequently asked questions

How long does it take to turn a photo into a video?

It depends on the selected model, video length, resolution, and provider load. Generation runs on the server; you can follow the task status and download the result when it completes.

Do I need video editing experience?

No. You do not need a timeline or editing software. Upload an image, write a motion prompt, choose the supported settings, and generate.

Can I use the videos commercially?

Check the Genjutsu AI terms and the applicable model or provider licence before using generated footage in ads, product listings, or client work. Commercial rights are not guaranteed merely by purchasing credits.

Which image formats are supported?

JPG, PNG, and WebP are supported by many available models, with a 12 MB per-image upload limit on the configured video workflows. Some models accept only JPG or PNG; the uploader checks the model you choose.

Why does my result look distorted?

The subject may be too close to an image edge, the input may be small or blurry, or the requested movement may be too aggressive. Try a sharper image, a wider crop, and a slower, simpler camera move.

Are free videos watermarked?

Watermark behavior can depend on the model and provider. Check the generated file and the applicable terms before publishing; this page does not promise watermark-free output.

Prefer a fixed scene instead of writing your own motion prompt? These Genjutsu pages lock a ready-made look and still start from a photo.

Ready to animate a photo?

Your next clip can start from one good photo.

Upload an image, write a short motion prompt, and compare the credit estimate before generating. If you need more credits, choose a pack that fits your plans — you do not need to buy a subscription just to open the editor.

Review the credit estimate before generating