form
Video to prompt generator
Upload a frame from a video, describe the motion, and get a detailed AI video prompt to copy.
Your inputs
markdown outWhat you get
Run lands herePublish
Share page, result, or card
This video to prompt tool works from a single frame. You upload one screenshot from your video - not the video file itself - and describe in a sentence or two what moves. It then writes a detailed AI video prompt covering subject, setting, lighting, color, shot type, camera movement, motion and mood, plus a short version, an avoid list and three variations.
How it works: a frame, not a video file
The tool reads still images only. It cannot open, play or analyze a video file or a link. That is why it asks for one frame and a short description of the action. The frame gives it everything about the look of the shot: the subject, the location, the light, the colors and the framing. Your description supplies what a still cannot show - how the camera and the subject move. The model describes only what is visible plus the motion you wrote, and does not guess at the rest of the clip.
How to grab a frame from your video
Pause the video on the moment that represents the shot and take a screenshot. On Windows, press Windows + Shift + S and drag over the player. On a Mac, press Shift + Command + 4. On a phone, pause and use the normal screenshot buttons, then crop out the player controls. Choose a sharp frame without motion blur, subtitles or on-screen menus, and play the video at its highest quality first so the details are clear.
How to describe motion
Cover three things in plain words. First the camera: static, slow push in, pull back, pan left or right, tilt up, tracking alongside, handheld, or orbiting the subject. Second the subject: what they do from start to finish, such as 'turns her head and smiles'. Third the environment: rain, drifting smoke, leaves, traffic, moving light. Add the pace and rough length, such as 'slow, about five seconds'. If you leave the box empty, the tool proposes simple motion and labels it as suggested.
Text-to-video and image-to-video prompts
A text-to-video prompt has to describe the whole scene, because the model starts from nothing. An image-to-video prompt starts from a picture you supply, so the look is already set and the prompt should mostly describe movement and camera work. Pick the target in the dropdown and the video to prompt generator adjusts the balance. The same frame also works as a video prompt from image input when you want to animate a still photo.
Questions
Can I upload a video file?
No. The tool accepts one image of up to 8 MB. Take a screenshot of a frame from your video and upload that, then describe the motion in the text box.
Which frame should I choose?
Pick the clearest frame that shows the main subject and the setting, usually near the middle of the shot. If the video has several different shots, run the tool once for each shot.
Does it recognize people in the frame?
No. It does not identify anyone by name. People are described by what they wear and what they do, which is also what a video model needs.
Will the prompt recreate my video exactly?
No. A prompt describes the style and action of a shot, and every video model interprets it in its own way. Use the variations to try different camera moves and lighting.
Is Video to Prompt free?
Yes, it is free to use. Only use frames from videos you made or have permission to use as a reference.