HiAPI
Showcases

Cinematic Language Visual Notes: Reference-Image-Driven Shot Studies

Using Vidu Q4 Preview's reference-to-video mode to create a set of "cinematic language visual notes": first use reference images to lock in a consistent scene and subject (the same street corner, the same character, and the same lighting and shadow foundation), then use prompts to describe different camera behaviors—push in, pull out, pan, truck, tracking shot, orbit—so that the same frame is performed repeatedly under different camera movements, forming shot-study material that can be compared side by side. In terms of composition, the reference image plays the role of "scene consistency," while the prompt plays the role of "shot grammar": in the prompt, refer to the material in the order of reference image 1, and clearly specify the camera path, speed, and changes in shot size; keep duration within the short-film range to maintain a single shot intention, and use the default settings for resolution and aspect_ratio to facilitate horizontal side-by-side comparison. Applicable scenarios: shot communication during storyboarding and previsualization, teaching examples for cinematography and camera movement, camera setup pitches for commercials and short films, and occasions where you need to explain to non-production parties the difference between "different ways of shooting the same scene." Because this is a Preview version, capabilities and available parameters will continue to iterate, so it is recommended to use it within the range declared by the current schema.

Style Camera Movement · Storyboard · film-grammar

Model Vidu Q4 Preview

Reference inputs

  • reference_image_urls

    reference_image_urls

Prompt

Try this model
At dusk, a rain-slicked, reflective city street corner, a single continuous shot, neon signs mirrored in shallow puddles. A solitary figure in a dark coat stands waiting under a metal awning. The camera performs a deliberate camera movement: a slow push-in from a wide establishing position to a medium close-up of the character's face, keeping the horizon level, with slight parallax in the background. Soft, motivated sign lighting, cool cyan shadows contrasting with warm amber highlights, fine film grain, shallow depth of field, 35mm anamorphic widescreen texture. Steady pacing, no cuts, no whip pans, restrained documentary tone.

Copy the full prompt and adapt it for your own generation.

Tags