Seedance 2.5 dolly zoom AI prompt for a horror shot
A Seedance 2.5 video prompt for a dolly zoom, the Vertigo effect, in a short hallway horror scene with a mannequin.
- Model
- Other
- Task
- Video scene
- Source
- Shared by @MrDavids1 on X
Travis Davids (@MrDavids1) posted this on X on October 2, 2026, in a series on film camera moves that AI video generators can copy. The prompt opens with the move itself: the camera dollies backward while the lens zooms in, so her face holds its size while the room swells around her. Keep that opening sentence and change the scene. The author’s tips: centre the subject against converging lines such as corridors or aisles, keep the move slow, and use it once, at the emotional peak.
The prompt is quoted from Travis Davids’ post on X.
- Prompt
- The camera begins a dolly zoom, dollying backward while the lens zooms in, the face stays fixed in frame while the room slowly swells and presses in around them A terrified women is standing in a hallway, standing still. Behind her a mannequin turns its head to look at her. Suddenly the lights flicker and turn off and then quickly turn back on but the mannequins are no longer present in the room.
- What it produces
- Asks for a horror clip in which the hallway swells around a frightened woman's face while a mannequin behind her turns and then vanishes when the lights flicker.
- Parameters
- Seedance 2.5 on Dreamina, image reference
- Source link
- x.com
- Original post
- x.com
- Added
Questions
What is a dolly zoom?
Which way should the camera move?
Which video model did the author use?
See also
Sources: each prompt is quoted from the official example or the public post linked on its page. Promptline is not affiliated with OpenAI, Anthropic, Google or Midjourney.