Grid groups

In a plain group the nodes sit wherever you drop them, and position is only about how it looks. DOTV's grid group is different: each image takes one cell, and the cell it sits in is itself information — the image in cell 3 is the top-right piece of the source image. The batches you get so often in AI video creation — a nine-cell storyboard, a set of angles for comparison — land on the canvas as grid groups. This section covers where they come from, how one image per cell works, and the toolbar buttons.

Where they come from

Grid groups are mostly built for you by something else, but you can also turn a set of images you already have into one:

Where from	What you get
Split grid in the image toolbar (see 3.5)	One image cut N×M, one piece per cell
Upscale image in the image toolbar	Cut up, upscaled cell by cell, then stitched back together
Generate storyboard on a script node (see 6.6)	A storyboard set with one shot per cell
To storyboard group on a plain group's toolbar (see 2.3)	The images you already have, laid out on the squarest grid that fits them; the spare cells stay open to fill

One image per cell

The cells are fixedEach image takes one cell, and the cell's position is itself information (which part of the source it maps to). Drag an image inside the group and it snaps to the nearest cell; if that cell is taken, the two swap.

1 / 6

A 2×3 grid group: four cells hold images, the other two are dashed boxes with a "+".

Drag an image inside the group and it snaps to the nearest cell when you let go. If that cell already holds an image, the two swap (position is information here, so pushing one of them aside would make no sense).
To pull one image out on its own, drag it outside the group and let go: it turns back into a plain image node on the canvas and no longer holds a cell (the group stays, with that cell now empty).
An empty cell is a dashed box with a +: click it to Upload an image or Pick from history and fill that cell.
A grid group has no resize handles — its size is worked out from rows × columns × the per-cell ratio, so any size you drag out is gone at the next reflow. To change the size, change the grid or the ratio.

The toolbar buttons

Shrinking the grid does not delete images. Going from 3×3 to 2×2 moves the images that fall outside the new grid into whatever cells are left; the ones that genuinely do not fit stay in the group without holding a cell. To actually delete them, hit Clear yourself.

Select a grid group and the toolbar swaps to a different set (Arrange and Run group from a plain group mean nothing here: positions are pinned by cell, and what is inside are finished images):

Button	What it does
Ratio	Changes the picture ratio of every cell; the group reflows to the new ratio.
Grid N×M	Changes the number of rows and columns.
Stitch	Stitches the images back into one large image by cell position (4k on the long edge), dropped next to the group as a new image node. Greyed out when the group holds no images.
Numbers	Marks each cell with its position number — for checking the order before you stitch. Click again to hide them.
Clear	Deletes every image in the group, keeping the cells so you can fill them again. Asks first.
To plain group	Drops the grid layout, after which the images can be arranged freely and the toolbar goes back to the plain-group set.

Ungroup and Download all are on both toolbars and work the same as on a plain group (see 9.2).

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

Open DOTV Download the desktop app

9.2 Grouping

9.4 Running a whole group

Grid groups

#Where they come from

#One image per cell

#The toolbar buttons

Where they come from

One image per cell

The toolbar buttons