Animate a still image with controlled movement. Upload a starting frame, add an optional ending frame, and describe how the shot should move.
Start frame
End frame
Fill in the form on the left and click "Generate" to create your own video.
Download links expire in 24h — please save your videos promptly.
View My Video History →From one still frame to controlled motion
With image to video, the model can see the subject before it invents movement. Wardrobe, palette, framing, facial features, product color, and background layout begin with a concrete reference. This makes the workflow useful when a creative team already has approved photography or artwork and wants to explore motion without describing every visual detail again in text.
A subtle head turn, fabric movement, light sweep, product rotation, rising steam, or slow camera drift can be enough to transform a still. Image to video does not need dramatic action to feel alive. Modest movement often protects identity and object shape better, gives the viewer a clear focal point, and produces a clip that is easier to combine with typography or editing.
Seedance can use a required starting frame and an optional ending frame. The start establishes the composition, while the end can suggest where a person, camera, or object should arrive. Image to video still has to invent the path between them, so choose compatible images and keep the transition readable rather than asking for a complete scene change inside one short generation.
The image is not the whole instruction. Tell image to video which elements should remain stable: face shape, hairstyle, package label, jewelry, room geometry, poster text, or the position of a key prop. Then describe one action and one camera behavior. This division of jobs helps the model understand what belongs to the source and what it is allowed to animate.
A reliable motion plan
A good image to video workflow starts before generation. Choose the strongest source, plan the first visible change, and decide which details are protected. After the result arrives, review motion and identity separately so a smooth camera move does not distract you from a face, label, or object that changed along the way.
Choose a sharp image with a clear main subject and readable edges. Image to video is better at adding movement than repairing weak composition, missing limbs, compressed text, or an object cropped at the place where it needs to move. Use the highest-quality source you have, remove accidental clutter, and leave open space in the direction of travel or camera movement.
Write the first three seconds as a visible action: the subject turns, steam rises, fabric moves, a bottle rotates, curtains shift, or the camera pushes toward a detail. For image to video, separate subject motion from camera motion. A product can turn while the camera stays locked, or the subject can remain still while the camera moves. Combining both is easier when each request remains gentle.
Watch once for timing and once for consistency. Image to video drafts can feel polished while small text, hands, jewelry, reflections, seams, or facial proportions drift. Pause the first, middle, and final frames, compare them with the upload, and reduce motion when the model changes the parts that make the subject recognizable. Save the prompt that protected those details best.
Direct motion without losing the still
Begin by naming the visual details that should survive the entire image to video clip. Identity words are more useful when they are concrete: same face shape, same hairstyle, same jacket, same product label, same room layout. Avoid asking to preserve everything, because that gives the model no priority. Tell it which three or four details make the source recognizable and which background areas may move more freely.
Describe motion with a physical verb and a direction. A person turns toward the window, a hand lifts the cup, a camera pushes slowly forward, or light travels across the package from left to right. Image to video responds better to a simple, ordered action than to broad instructions such as make it dynamic. If the clip needs several beats, generate them separately and keep the approved source frame consistent between shots.
Use negative direction sparingly and clearly. Phrases such as no camera orbit, no extra objects, static wall, or keep the label unchanged can protect a commercial brief. A long list of warnings may compete with the positive motion instruction, so keep image to video constraints close to the element they govern. State the desired movement first, then add the few limits that prevent the most costly failure.
Where a still image becomes useful motion
Image to video works best when you already like the still image and only need one clear movement for an ad, post, storyboard, or landing page.
Turn a packshot into a slow rotation, pour, light sweep, or camera reveal. Keep the source clean and check labels before publishing.
Add a small head turn, blink, breath, or expression change. Use a clear face and keep the movement simple.
Give a room or landscape a slow camera push, moving light, clouds, water, or distant activity for a more alive scene.
Add parallax, camera drift, light movement, or one animated detail. Keep important lettering static or replace it in final editing.
Clear answers
Clear answers about choosing a source image, writing motion prompts, protecting identity, and reviewing an image to video generation.
Image to video turns a still photo, design, or illustration into a short moving clip. You upload a starting frame and describe the subject motion, camera, timing, atmosphere, and details that should stay close to the source. Seedance creates the new frames between the beginning and end of the requested action.
Use a sharp image with one clear subject, readable edges, useful resolution, and enough space for movement. Image to video performs better when the person or object is not heavily cropped or hidden. Clean hands, faces, product shapes, logos, and background lines give the model a stronger first frame to preserve.
Start with the details that must remain unchanged. Then describe one subject action, one camera behavior, environmental movement, timing, and optional sound. A good image to video prompt reads like a short shot direction. Use physical verbs, put events in order, and avoid asking for several dramatic transformations in one clip.
A clear portrait and modest motion can improve consistency, but faces, expressions, hair, hands, and body proportions may still drift. Review every image to video result at full size. When identity is critical, reduce movement, use a shorter duration, add explicit protected details, or move to a multi-reference workflow with consented images.
Yes. Product turns, pours, steam, condensation, light sweeps, and simple reveals are common uses. Hard edges, tiny labels, and packaging text can change during image to video motion, so use a clean source, keep the action restrained, and replace exact typography during editing when commercial accuracy is required.
Use image to video when one main still should define the shot. Use reference to video when separate sources need to control identity, product shape, location, movement, style, or audio. Start with the simpler workflow; add references only when the single image and prompt cannot explain the creative brief clearly enough.
Five to eight seconds is a useful first pass because a single movement remains readable. A longer image to video duration can work for a slow camera move or continuous environmental action. If the source begins to drift near the end, shorten the clip, reduce the movement, or provide a compatible ending frame.
Commercial use depends on your plan, provider terms, and your rights to the uploaded image and depicted subjects. Use owned or licensed sources, get consent for recognizable people, verify product and factual claims, and review the current terms before publishing image to video output in advertising, client work, or paid media.
Choose a strong source frame, describe one controlled movement, and protect the details that make the image recognizable. The image to video generator at the top of this page is ready for your first motion test.
Create image to video