To get the most out of video generation, you should always seek to include as much detail as you can about the motion you want to achieve. This should include:
Focal Point: What is the main subject of the shot? Specify the key focus of the scene.
Camera Settings: Type of lens, focal length (wide, standard, or telephoto), aperture settings for depth of field.
Type of Shot: Close-up, wide shot, over-the-shoulder, etc.
Camera Movements: Pan, tilt, zoom, dolly, tracking, etc.
Scene Description: Environment, time of day, weather, and background details.
Lighting: Natural, soft, hard, backlight, or colored lighting.
Mood/Tone: What atmosphere are you trying to convey? (e.g., tense, serene, mysterious).
Frame Rate: Specify if the shot should feel slow-motion or fast-paced.
Color Grading (if applicable): Muted, vibrant, monochrome, etc.
You should also always follow our top tips for video prompting:
Be Clear and Specific: Detail each aspect of the scene and camera movement to get the desired effect.
Use Action Words: Especially for camera movements (e.g., “panning,” “tracking,” “zooming”).
Think in Visual Layers: Start with the big picture (environment), then layer in the details (subject, movement, lighting).
Match the Lighting & Mood: Sync the lighting with the tone you want to achieve (e.g., soft lighting for a romantic scene, harsh lighting for tension).
Test and Iterate: Runway's models respond well to refined input, so test your prompts and tweak them to achieve your vision.
To find out more about differen camera shots and movements, click here.