Photo to Video for Social Media: Vertical Clips and Safe Crops

Short answer: choose the platform and the crop before you generate, keep the subject centered with room at the top and bottom, ask for one small movement, and add captions in an editor after export. Do not ask the video model to write text inside the frame. Review the paused frames before you post.

A vertical video is not just a landscape video with the sides cut off. The subject has to sit inside a tall frame, the movement has to make sense at that scale, and the platform’s interface will cover parts of the screen. Planning the crop first saves a lot of re-generation.

Choose the platform and crop first

TikTok, Reels, Shorts and Stories all use a vertical frame, but the safe areas differ. The top and bottom of the screen may hold captions, buttons, a profile name or a progress bar. Keep the face and the product in the middle third, and leave breathing room around them.

If the original photo is landscape, decide whether you are cropping to a vertical frame or adding a background. Cropping can cut off a hand or a product edge. Adding a background changes the composition. Choose one before you animate, not after.

The free demo does not expose a reliable aspect-ratio selector. If you need a specific 9:16 output, use the signed-in image-to-video tool. The paid tool shows the supported formats before you spend credits.

The safe-zone checklist

Before uploading, mark the area that must stay visible:

• the face and eyes;
• the product label or the main object;
• the hands, if they are part of the story;
• the top of the head;
• the bottom of the frame where captions may sit.

If the subject is too close to the edge, the model may crop or bend it during a camera move. Leave more space than you think you need. A centered subject with room around it is easier to animate and easier to caption.

Do not put important text in the generated frame. The model may redraw it, misspell it or dissolve it during motion. Add captions and titles in an editor after export.

One motion that fits the format

For a vertical portrait, a fixed camera with a blink or a very small head turn is usually enough. For a product, a slow rotation or a push-in works if the object stays inside the safe area. For a landscape, a slow vertical tilt or a push-in can fill the tall frame without a lateral move that reveals missing sides.

Avoid a fast pan or a big zoom. Vertical video is watched on a small screen, often without sound. A small, readable movement is more effective than a busy camera path that makes the subject leave the frame.

Captions, music and export

Add captions after the clip is exported. Use a readable size, a strong contrast and a safe margin from the bottom edge. If the platform adds its own interface, keep the captions above it.

Music is also a post-production step. The free animation demo focuses on motion and does not provide a reliable audio track. Add music or voiceover in an editor, and check the platform’s licensing rules.

Export at the highest resolution the tool provides, then compress for the platform if needed. Do not upscale a low-resolution clip and expect it to look sharp. Start with a clear source photo and a clean render.

A review loop for social clips

1. Watch the clip on a phone-sized screen.
2. Check the first frame and the frame with the most movement.
3. Confirm the face, hands and product label are inside the safe area.
4. Add captions and music in an editor.
5. Watch the final export once more before posting.

If the clip only works on a desktop monitor, it is not ready for a phone feed. The small screen exposes cropping, soft focus and text problems that are easy to miss on a large display.

Keep a small library of your best-performing crops and motion recipes. When a format works, repeat the structure with a different photo instead of starting from a blank prompt every time. That is how a one-off test becomes a reliable posting workflow.

What to do when the clip is too short

A short clip is not a failure. Many social formats reward a tight loop or a three-to-five-second moment. Trim the clip to the strongest segment and use a clean cut or a crossfade. If you need a longer story, combine several clips rather than asking one model to generate a long, complex scene.

For a loop, keep the camera fixed and the movement small. Trim the ends in an editor so the last frame can return to the first. The free demo does not guarantee a seamless loop.

FAQ

Can I choose 9:16 in the free demo?

The free demo does not expose a reliable aspect-ratio selector. Use the signed-in image-to-video tool when a vertical frame is required.

Where should I leave safe space?

Keep the subject centered and leave room at the top and bottom for platform UI. Avoid placing a face or product label near the extreme edges.

Should I add captions inside the generated video?

No. Generated lettering is unreliable. Add captions in an editor after export so they stay readable and can be updated per platform.

How long should a social clip be?

Start with the shortest clip the model can produce and trim in an editor. A tight three-to-five-second loop is easier to review than a long clip with a weak ending.

Make a vertical test

Start with the no-login photo animation to check whether the source and the motion work. When you need a true vertical frame, move to the signed-in image-to-video tool, and compare the credit packs before you commit to a plan.