as the title describes... i would like to have depth information in the alpha layer of the camera (both the photo and the stream camera) so we can use this data in photo editing and video editing software (effectively creating a mask based on depth, instead of transparency/chroma key)

my use case would be to send only the foreground(WHITE) / background(BLACK) to a compositing software (resolume, synesthesia, photoshop, premiere...)

you could also use this with premultiplied alpha for a performant dynamic masking solution (we use this a lot with compositing video sotftware...)

Picture for example, we want the bottom right quadrant (white is foreground, becomes opaque closer to 1) which would work very nicely for pre-multiplied alpha setups