Image to video aspect ratio: prepare the frame before the motion.

Choose the video shape first, preview the matching crop, and check that the subject has room to move. Turning a landscape photograph into a vertical video can remove much more of the scene than you expect.

1. Choose a shape, then choose a resolution

Aspect ratio is width divided by height. A 16:9 frame is wider than it is tall; 9:16 is taller than it is wide; 1:1 is square. Resolution is the number of pixels. For example, 720 × 1280 and 1080 × 1920 both have a 9:16 aspect ratio, although they contain different amounts of detail.

Decide where the finished clip will sit in your edit. A vertical composition gives you height for a standing person or tall object. A wide composition gives two subjects or a broad environment more horizontal space. If you need both versions, plan two crops from the source before generating either clip.

Check which ratios the selected model and mode actually support. Google's current Veo image-to-video console instructions offer 16:9 and 9:16. That is a provider-specific setting, not a promise that every video generator accepts any shape you upload.

2. Preview how much of a landscape image survives

Suppose your photograph is 1600 × 900 pixels and you want a 9:16 frame filled edge to edge. Keeping its full height means keeping a width of 900 × 9 ÷ 16 = 506.25 pixels. You retain about 31.6% of the original width. A centered crop removes roughly 547 pixels from each side, so a second subject near an edge may disappear entirely.

Geometry examples, not generated-video tests. Pixel rounding varies by editor. These figures assume cropping to fill the target shape without stretching; a generator may handle the input differently.

To calculate another crop, compare the source width-to-height ratio with the target ratio. If the source is wider, retained width equals source height × target ratio. If it is taller, retained height equals source width ÷ target ratio. Position that rectangle over the details you need before exporting the reference.

3. Choose between cropping, fitting, and reframing

Crop when the important details fit

Cropping removes outer content while keeping the remaining subject proportions. Move the crop toward the subject rather than assuming the center is correct. Check hands, feet, product edges, and any object the subject will interact with. Export from an image editor at a size your chosen generator accepts.

Fit when you need the whole photograph

Fitting scales the complete image proportionally inside the target canvas, leaving unused space. A wide photograph inside a vertical canvas leaves space above and below. You can intentionally design those areas in an editor, but do not assume the video model will preserve a border exactly. If borders must remain fixed, add them around the finished clip during editing instead.

Reframe when a crop cannot preserve the shot

Take a new photograph with the intended composition, reposition the subject in a layout, or use another accurate reference. An image expansion tool can propose new surrounding content, but inspect those invented areas carefully, especially near people, product edges, or text. More canvas does not recover details the original never showed.

4. Leave space for the movement and the edit

A subject that fits at the opening may leave the frame after a step or camera move. Check the direction of travel and the intended ending. A push-in makes the subject occupy more of the frame; start with enough surrounding room or use a smaller move. Leave space for titles you intend to add later, then preview those titles in the actual edit.

This original planning exercise has not been rendered or tested. Use it with a reference that already includes the whole subject:

Set 9:16 in the generator separately if that is your deliverable. Review the full result: a request to hold composition is a useful brief, but does not guarantee a perfectly fixed frame. For product references, use our product photo to video review checklist to inspect shape and packaging as well as the crop.

5. Confirm the input handling before generating

Runway's current Gen-4.5 guide says the aspect-ratio control accommodates the input image by default and can crop it when changed. Inspect that control and its preview rather than relying only on the dimensions of your uploaded file.

1. Confirm the model, image-to-video mode, and supported output ratio.
2. Inspect the uploaded reference after any crop or resize step.
3. Check the whole subject, essential edges, and room for movement.
4. Save the exact reference, prompt, ratio, and resolution with the attempt.
5. After rendering, check the downloaded file dimensions and the opening, middle, and ending frames.

Draft the movement and framing notes in Shotmelo's image-to-video prompt planner, save the local draft, and export the text plan for your chosen service. Prepare the crop in an image editor. Shotmelo's current preview does not crop images automatically or render videos.

Common framing questions

Does writing “9:16” in the prompt change the video size?

Use the output-ratio setting when available. Text about a vertical composition can explain your intention, but does not replace a supported size control. Confirm the dimensions of the actual output.

Can I resize a landscape photo directly to 1080 × 1920?

Forcing those dimensions without preserving proportions stretches the photograph. Crop or fit it proportionally instead. Increasing the pixel count also does not restore detail missing from the source.

Should I crop the source or the finished video?

Prepare the source crop when it helps establish the desired opening composition. Crop the finished clip when you need an alternate edit, but review every frame: the subject may move outside the new crop. Exporting a cropped version can also leave fewer pixels to work with.

Choose a source and motion plan together. If you are still selecting a generation service, read our AI video generator selection guide for the workflow questions to check before spending on an attempt.