Text2Movie

The film language each AI video model's docs let you write

Directing note: moves from references

A few models take a camera move from something other than words: a reference video, a single image or a drawn path. As of 2026-09-25.

Moves from referencesThe note in 4 steps: Reference, Path, Image and Result.ReferenceThe decisionat this point.thenPathWhere modelsdiffer.thenImageWhere modelsdiffer.thenResultWhat the docsoffer here.
Fig. 1 Left to right, in the order the note takes.

That suits moves that are hard to describe, and it ties the result to something a director can show.

It also makes the move depend on the reference being clear.

1From the docs

use the Motion reference option to upload a video and extract its camera movements, including pans, zooms, tilts, and motion paths, which will guide the generation.

Adobe, Firefly help: generate videos using text prompts

Reference character, motion, camera, style, voice, or editing rhythm

MiniMax, video generation guide

Create cinematic camera moves using just a single image.

Moonvalley, Marey

2Camera language in a pipeline

Where “Moves from references” has to be written by hand, SceneMixer's pipeline writes the shots; its guides explain:

SceneMixer turns a novel into a watchable short-drama series in one pipeline: upload the novel (Chinese or English) → AI extracts characters, scenes, and an episode-by-episode structure → generates consistent character and scene art → writes storyboard scripts → renders video segment by segment (16:9 by default, 9:16 in one setting) → composites full episodes.

SceneMixer, Turn a Novel into a Video Series with AI (read 2026-10-13)

On this site it is the 720P tier of the family; the storyboard, references and voices are the same as on the other two tiers.

SceneMixer, Seedance 2.5 Lite — “Is Lite a lower-quality model or a lower resolution?” (read 2026-10-13)

3Other notes

Examples are quoted from the models' docs; no model was run.