How to animate a photo with artificial intelligence

TechSkill Editorial

TechSkill Blog

October 19, 2026 - 5 min read

Key Takeaways

    To animate a photo with artificial intelligence, upload an image to an image-to-video tool, describe one simple movement and create a short clip first. For example, ask the leaves in a photo of a potted plant to sway gently while the camera stays still. Then check whether the plant's shape or background changed during the motion.

    Do you need photo animation or a slideshow?

    First decide what should move. If you want a slow zoom into one photo or several photos shown in sequence, a video editor can do the job. Canva's photo video editor offers photo slideshows and MP4 downloads. (source 3).

    If you want an object inside the image to move, try image-to-video generation. The movement is synthesized. A video made from an old photograph is not a recording of movement that happened in the past.

    For this exercise, imagine using a photo you took of a potted plant. The task is a five-second vertical video: the leaves sway gently while the pot and camera remain still. This is a learning example, not a description of a video already generated.

    Choose a photo with clear details

    Open the photo at full size before uploading it. Check whether the leaf outlines are distinct or overlap confusingly. A cropped leaf, blurred edge or complex background makes motion harder to evaluate.

    Choose an image you have permission to use. Keep the main subject close to the center and leave some space around the edges. For a vertical video, inspect the 9:16 crop: the top of the plant and the pot should remain in the frame.

    Runway's guide explains that the starting image already establishes composition, lighting and visual style. Use the text prompt to describe motion. (source 1).

    Create the first video with one movement

    In Runway's Gen-4.5 web interface, choose Image to Video and upload the photo. The model supports videos of 2–10 seconds; select 5 seconds and a 9:16 aspect ratio for this exercise. The official guide lists a Standard or higher plan as the access requirement. Check availability and cost in your account before generating. (source 2).

    You can begin with this short prompt:

    The plant's leaves sway gently in a light breeze.
    The camera remains still. One continuous shot.
    

    The prompt asks for gently swaying leaves, a stationary camera and a single continuous shot. Choose duration and aspect ratio in the separate settings.

    Do not add camera rotation, a lighting change and a new object appearing all at once on your first attempt. A simple request makes it easier to identify the cause of an unwanted result. A stationary camera is a request; you still need to check whether the generated video follows it.

    What should you look for when reviewing the video?

    Watch at normal speed first, then pause at the beginning, middle and end. For the five-second example, compare roughly 0, 2.5 and 5 seconds. If you cannot pause at exactly five seconds, inspect the last available frame.

    What to check Acceptable result When to try again
    Leaves Sway gently Merge together or new leaves appear
    Pot Keeps its shape and color Stretches, rotates or changes color
    Background Stays largely unchanged Gains a new object or flickers sharply
    Camera Holds a stationary frame Zooms or moves sideways unexpectedly

    If the leaves move too strongly, soften only the motion description on the next attempt. For example, replace the first sentence with:

    The plant's leaves move very slightly in a faint breeze.
    

    Keep the remaining camera and continuous-shot instructions unchanged. When comparing the two results, you will know exactly what you changed. If shapes still break down, try a simpler photo too.

    Finish the clip in an editor

    Download an acceptable result and prepare it in a video editor. Trim an unstable beginning or ending if needed. If you loop the movement, check for a jump from the last frame back to the first.

    You can add Georgian instructions or narration through text-to-speech. Save the source photo, the prompt and the selected video so you can continue from the same starting point later.

    If preserving the exact shape is essential and generated movement changes the details, choose a conventional photo zoom or slideshow. Let the task guide the choice: the video should preserve the details that matter to you.

    Full source addresses

    1. https://help.runwayml.com/hc/en-us/articles/48324313115155-Image-to-Video-Prompting-Guide

    2. https://help.runwayml.com/hc/en-us/articles/46974685288467-Creating-with-Gen-4-5

    3. https://www.canva.com/create/photo-videos/

    Updated: October 19, 2026

    Subscribe to the TechSkill Blog

    Discover where product building is heading next

    Share this post

    By sharing your email, you agree to our Privacy Policy and Terms of Service

    TechSkill

    This content will be available here soon.

    Back to courses