Depth Pass
Make a depth map and a normal map from a picture or a clip, and what to do with them.
Depth Pass makes a greyscale depth map from a picture or a clip. Each shade shows how near or far that part of the frame is. It also makes a normal map, worked out from the depth.
The depth is scaled once across the whole clip, so it stays steady from frame to frame. It shows what is nearer and what is farther, not distances in metres.
- On the Shots page, open a take, press Transform and choose Depth Pass. Or, in the Graph, wire the picture or clip into an Edit Video node and set its workflow to Depth Pass.
- Choose a quality. Stable gives the steadiest result. Fast is quicker, with a little more drift on a long clip.
- Press Run and wait for the result.
- In the Graph, the Edit Video node lists both maps when the run finishes, with a download button for each format. From a take's Transform menu, the result arrives as a new take on the shot.
- From a clip: an MP4 of each map, and a ZIP of PNG frames for each.
- From a picture: a PNG of each map.
- Each map is the same size as the source. Clips keep the source frame rate and have no sound.
Bring the depth map into your compositing or editing software:
- Fog and haze: Use the depth map as a matte so fog or haze gets thicker with distance.
- Focus: Feed the depth map to a lens blur to soften the background, or move the focus after the shot.
- Layering: Cut a matte from the depth map to put a title or an element behind your subject and in front of the background.
- Grading by distance: Grade the background apart from the foreground, for example to darken it or cool it down.
If your software expects near and far the other way round, invert the map first.
Fine detail like smoke, sparks or fire may be missed. The depth map is 8-bit, so a strong blur or thick fog can show steps. Edges are soft on large frames, because the depth is worked out at a smaller size first.
Still stuck? Support has the common problems and a way to reach us.