Guide
English
Search the docs⌘K
8.1

Previz nodes

Some shots are impossible to put into words. "From behind her left shoulder, waist height, tilted slightly up, with a railing in the foreground" — write that ten times and the model still may not give you the angle; run it again for the next shot and the people and the set have drifted apart again. DOTV's Previz lets you lay it out instead: place the people, the props and the camera in a 3D space, then take a picture from the camera's point of view. That picture is a gray-and-white blockout, not a finished frame — its job is to be a composition reference: wire it into an image node so the model paints from that angle, with that blocking (see the reference images in 3.2). The camera positions and blocking that prompts can never quite pin down in AI video creation get set exactly right here, once.

When to use it

Three things it saves you over writing prompt after prompt:

What you want How you do it here
Several camera angles on one scene Set the scene once, put three or five cameras in it, shoot them all at once
Multiple angles on the same person Place one person and orbit around them for twelve shots (see 8.7)
Blocking and relative positions you cannot describe Just put the people and the props where they belong and let the picture say it

What the card looks like

The previz nodeThe card makes no content of its own — it holds a 3D scene. To block a shot or capture one, hit Open previz and go full screen.
1 / 2
The image on the left feeds it a panoramic backdrop; the one on the right is the previz node.

Open previz goes full screen — setting the scene, placing cameras and taking pictures all happen in there (from 8.2 on). Once a version has been built, the card shows the scene itself: whatever angle you left the previz stage at is the angle on the card, with the button floating in the middle of the picture.

An image node wired in from the left is the Panorama background: once you are inside, that image wraps around the scene instead of an empty sky. Wire in several and only the first one is used; wire in none and it works just the same.

Selecting the card pops up an input bar along the bottom: describe the scene you want ("outside a convenience store, dusk, two people talking"), and it builds it out of what the previz stage has — mannequins, primitive shapes, cameras — then you open the previz stage and keep editing. Sending another line after that rebuilds the whole thing (it overwrites the current version). The Reference chip picks an image from the canvas: it guides the build and doubles as the previz stage's panorama background — "build from this image" and "use it as the backdrop" are one and the same wire.

This card holds that scene (how many people you placed, where the cameras are, which pictures you took), so:

  • Close the previz and open it again, and the scene is exactly as you left it — your viewing angle too.
  • The scene travels with the canvas — open the same project on another machine and everything you placed is still there.
  • Copying this node copies the whole scene, so you can rework a second camera setup without touching the first.
Try it in DOTVReading only goes so far: open DOTV and make an AI video with the feature this section covers.