Guide
English
Search the docs⌘K
8.7

Capturing a frame

Blocking the scene was for this step: capture what the camera sees and take it back to the canvas as a reference image. DOTV's Previz has three ways to shoot — one frame of the current view, one at each of four angles around the point you're looking at, or a tight ring of twelve — and the captures go straight onto the canvas, ready to be wired into an image node. This section covers the three ways, where the captures land, and what makes it into the frame and what doesn't.

Three ways to shoot

The three buttons at the bottom of the toolbar differ in how many frames they take, top to bottom.

Capture this viewShoots exactly what you see. The grid, camera wireframes and transform handles stay out of the picture; character labels don't — turn them off in the scene panel if you don't want them.
1 / 3
Three capture buttons: one frame, four, twelve.
Which one What you get When to use it
Capture this view The one frame you are looking at The composition is settled and this angle is the one you want
Capture 4 angles An even ring around the point you're looking at, one frame at each of four angles One scene, a few camera positions to choose between
Capture 12 angles The same ring, packed to twelve frames Multiple angles on one character or one prop for reference

The two ring captures only change the horizontal angle — the elevation and the distance stay as your current view has them — so none of the twelve come back pointed at the sky or the floor. For a ring that is higher or further out, set the view first, then press.

Where the captures land

A camera's capturesCaptures hang under the active camera (with no cameras at all they go straight to the canvas). Hover one for To canvas and Delete; the two at the top are All to canvas and clear.
Captures hang under the camera that took them.

Captures go straight onto the canvas, one image node each (the two ring shots arrive together as one storyboard grid). With an active camera (the one whose panel button reads Active), a copy is also kept in the camera panel, so you can place it again from there later.

Two buttons sit at the bottom right of every image:

  • Add to canvas — creates an image node on the canvas to hold it. This is the exit for the whole chapter: wire it into an image node as a reference so the model draws from this angle (see 3.2).
  • Delete — drops that one image only.

The two buttons at the top handle the whole set: Add all to canvas (everything this camera shot goes over) and clear.

What makes it into the frame, and what doesn't

A capture is not quite what you see in the viewport, and the differences are deliberate:

In the viewport In the capture
The lines of the grid floor Not included
Camera models, transform gizmos, the axes Not included
The name label over a character's head Included — turn it off under Toggles in the scene panel if you do not want it
The dimmed band outside the aspect frame Not included; the image is cropped to what is inside the frame (see 8.3)
Hidden objects (eye closed in the scene tree) Not included

So for a clean reference image, check two things before you press capture: that the aspect is right, and whether the character labels are off.

Try it in DOTVReading only goes so far: open DOTV and make an AI video with the feature this section covers.