Upload a reference image for canvas storyboards and clips
Give it an image, and it creates a storyboard and segmented video on the canvas. Great for short-video beginners.
Use this skill
What it is
This is a creative workflow that helps turn a still image into continuous video. It analyzes the image's color and lighting, plans multiple connected shots for you, and works to avoid style mismatches from shot to shot.
You only need one clear image, such as a product photo, landscape photo, or character poster. Send it to the AI on the right, then follow the prompts to choose your desired video type and style.
When it's done, a storyboard and multiple 15-second video clips appear on the canvas. You can edit and regenerate an individual shot, or arrange clips on the video timeline and export the final video.
An example
- You type
- Upload a still-life photo of a ceramic coffee cup and tell the AI you want a 15-second product ad showcase in a warm morning-light style.
- You get
- The canvas generates a storyboard card with 8 shots and corresponding 15-second video clip cards, carrying over the ceramic texture and soft lighting of the original image.
When to use it
- You have only one real product photo and want to make a short e-commerce showcase video with multiple angles.
- You've drawn a character or toy illustration and want to make a dynamic teaser for it.
- You took a beautiful landscape photo on your phone and want to expand it into a moody landscape short.
- You have a cinematic concept poster and need to plan shot setups and generate a sample video.
Problems it solves
You only have one still image and don't know how to plan shot transitions or visual content.
The AI analyzes the original image's characteristics and lists the shooting approach for each shot on the canvas so you can confirm the plan first.
The finished video's shots have different styles, making the visuals feel pieced together.
It designs new shots using the original image's materials, main colors, and lighting as references, trying to keep the visual style consistent.
One generated shot isn't what you want, and you're worried you'll have to start over from scratch.
Storyboards and videos are separate cards on the canvas, so you can edit the text for one segment and regenerate just that part.
How to use it
- In the skill library, click 'Use this skill,' then upload a reference image in the chat panel on the right side of the canvas.
- Choose the video type and style, and let the AI lay out storyboard cards on the canvas.
- Review the storyboard content, and once you're happy, have the AI generate the corresponding 15-second video cards.
- Review the video clips, arrange them in order on the video timeline, and export the final video.
Frequently asked questions
Can I generate a video without uploading an image?
No. This skill requires uploading at least one image first. The AI uses the image's lighting, colors, and subject as its reference; without an image, it can't start creating storyboards or videos.
Can the AI add voiceover or automatically edit the final video for me?
No. The AI can't add voiceover on its own, and it won't automatically assemble a final video. It generates separate video clips on the canvas; you can arrange them in order on the canvas video timeline and then export.
If I'm not happy with one generated shot, do I need to redo everything?
No. You can regenerate just the shot you're unhappy with. Each storyboard and video clip on the canvas is a separate card. Just edit that shot's text description and regenerate it—no need to start over.
Why is the finished video split into multiple short clips?
Because this skill plans and generates video shot by shot, making it easier to review and edit each part. Each storyboard card produces a corresponding 15-second video clip, which helps avoid visual style drift. After all clips are generated, you can arrange them in order on the canvas video timeline and export them together.