You are here: Reference Guide > Deep Nodes > DeepMerge

DeepMerge

DeepMerge lets you merge samples from multiple deep images, so that each output pixel contains all of the samples from the same pixel in each input. You can also use it to remove or fade out samples in the B input that are occluded by samples in the A input. Unlike the DeepHoldout node, DeepMerge doesn’t flatten the data but produces a deep output image.

Inputs and Controls

Connection Type	Connection Name	Function
Input	A, B	The deep images that you want to merge.

Control (UI)	Knob (Scripting)	Default Value	Function
DeepMerge Tab
operation	operation	combine	Sets how to merge the inputs: • combine - combine the samples from the A and B inputs. • holdout - hold out samples from the B input by the samples in the A input. This removes or fades out samples in input B that are occluded by samples in input A.
drop hidden samples	drop_hidden	disabled	When enabled, samples that are completely obscured by other samples are discarded (that is, samples behind other samples that have an alpha of 1). This control is only available when operation is set to combine.
drop zero threshold	drop_zero_threshold	0.0000001	Any samples whose alpha values fall below this threshold are discarded from the holdout. This can help filter out samples with very low alpha values caused by floating point inaccuracy. This control is only available when operation is set to holdout.