Any video on your board — one you generated in Layer or one you brought in yourself — can be used as a source for a still image. Pick a specific frame and Layer extracts it as a standalone image asset, ready to use anywhere a reference image is accepted.
How it works
Select the video asset on your board.
Choose the frame extraction option (available wherever cross-modality transforms appear for an asset).
Scrub through the video to find the exact moment you want.
Confirm — the selected frame is captured and added to your board as a new image, and saved to your Library.
💡 This is part of Layer's broader cross-modality support — the same way you can take a 3D model, view it from a new angle, and extract that view as a 2D image, you can pull a single frame out of any video.
Best for building longer, connected scenes
While you can extract a frame from any video for any reason, this is especially useful for chaining video generations together:
Generate a video clip.
Extract its last frame.
Use that frame as the First Frame of your next generation.
This carries visual continuity from one clip into the next, so a multi-clip sequence reads as one continuous scene rather than a series of disconnected cuts. See Creating videos with First Frame/Last Frame for the full walkthrough.
What's next
Use extracted frames to anchor a clip → Creating videos with First Frame/Last Frame
The full video generation walkthrough → Generating video on Layer
Bring in and manage reference images generally → Uploading assets and reference images