A strong still can do more than sit on a grid. The free image to video tool on CanvaPro animates a photo into a short clip, starting from the picture you already have rather than inventing a whole scene from scratch.
You upload a JPG, PNG, or WEBP up to 10 MB. A motion prompt is optional. If you want to steer the camera or the action, there is a text box for it, with room for up to 2,000 characters and a placeholder that suggests camera move, action, lighting, and mood. Complete the security check, click Generate Video, and download the clip when the queue finishes. The page does not show a preview of the video before the download is ready.
How this differs from text-to-video
Image to video keeps your photo as the base of the clip. The model adds motion — a slow camera push, wind in hair, water moving, light shifting — on top of that still. Text to video is a different job: it invents the whole scene from words, with no still required. If you already have a portrait, a product shot, or a landscape you want to bring to life, start here.
The result is a short animated clip, not a long edited film. Treat it as a looping social piece, a motion ad built from an existing campaign still, or a way to give a hero photo a little movement. The tool does not ask you to set a timeline, an aspect ratio menu, or a list of export presets on the page. You upload, optionally describe the motion, and generate.
When you have no photo at all, generate a short video from text instead of animating a still.
How to turn a photo into a video, step by step
- Choose a sharp still. Clear subjects with some depth — a person in a landscape, a product with a real background, a street or shoreline with space behind the subject — usually animate more believably than a flat screenshot or an already-blurred photo.
- Upload a JPG, PNG, or WEBP that is 10 MB or smaller. If a camera original is over the limit, export a slightly smaller copy and keep the full-size master elsewhere.
- Add a motion prompt only if you need one. This field is optional. Leave it empty if you are happy to let the tool invent a natural motion. If you want to direct it, describe the camera, the action, the lighting, or the mood in plain language, and stay within 2,000 characters.
- Complete the security check and click Generate Video once. Jobs are queued. Renders usually take a few minutes, depending on image size and server load. Leave the tab open and wait for the download to become available rather than submitting the same still over and over.
- Download the clip when it is ready. There is no on-page video preview on this tool, so open the saved file to review the motion before you post it. If the movement feels wrong, change the prompt or try a clearer still and run a new job.
Upload a sharp still, describe only modest motion, and download the clip when the queue finishes; a second try on the photo to video tool should use the same original still.
What to write in an optional prompt
Short, concrete directions work better than a long wish list. “Slow push in, hair moving in the wind, soft late light” tells the model what to do. A prompt that asks for a completely different scene, or for a new person who is not in the photo, fights the whole point of starting from a still.
You do not have to fill the 2,000-character limit. A sentence or two is enough when you have a clear idea. If you are unsure, skip the prompt on the first run, watch the result, and add direction only if the motion needs a nudge.
A short sound to sit under the clip can be made with the AI sound effect generator after you have watched the silent picture.
Stills that animate more cleanly
Portraits, products, and landscapes with a bit of depth give the model room to move a camera or suggest a breeze. Harsh, already-blurred photos and flat screenshots are harder, because there is less structure to animate. Start with the highest-resolution still you have, and avoid a file that has already been saved down several times.
Photographers often turn a hero still into a looping social clip. Sellers add subtle motion to a catalog shot. Marketers build a short motion ad from a campaign photo they already shot. In each case, the still should be a picture you have the right to use.
A packshot that should become a shop loop has a more specific path in how to turn a product photo into a short video.
If the job fails or the motion looks off
Check the format and the 10 MB limit first. If the queue seems slow, wait through one full run — a few minutes is normal — before you start a second job. If the clip does not appear in your downloads, look at the browser's download list and blocked-pop-up settings, because this tool is built to hand you a file rather than a preview player.
If the motion feels random, add a short prompt and try again with the same still. If the subject smears or the frame feels unstable, switch to a sharper photo with a clearer main subject. Keep the original still in your archive so you can generate another version later without hunting for the source.
When you do not have a still at all and you want a scene created from a written idea, use the text to video tool instead. That path starts from words, not from a photo.
If you need the person in the still to speak a line, that is a different tool, covered in how to make a talking photo from a portrait.
A short checklist
- Upload a clear JPG, PNG, or WEBP under 10 MB.
- Use the motion prompt only when you want to steer the camera or the action.
- Click Generate Video once and wait for the queue to finish — often a few minutes.
- Open the downloaded clip to review it, since the page does not preview the video first.
When you are ready, upload one still, add a prompt only if you need it, and save the clip before you start the next photo. One finished download is easier to post than several overlapping jobs.