Use ZDepth channel to make the Deep image.
Smoothing the harsh holdout intersection that occurs when the holdout itself doesn't have enough samples.
Deep2VP suite is a toolset to convert deep data to world space position, support single layer or volumetric of deep data. Now it has 5 nodes in total in this suite: DeepToPosition, DeepPositionToImage, DeepVolumeMatte, DeepVolumePattern, DeepVPCamProjection The zip included all toolset in gizmo and group format, demo script.
keymix for deep
DeepSlice use the deep channel to select a slice out of the input image.
A version of the DeepCrop node that allows the user to set a falloff for a soft transition. Can be used to gradually fade off anything that reaches a specific distance from the camera (like anything super close). This can for example be usefull to prevent visible 'popping' when a camera moves through geometry.
CopyBBox for deep image stream
It's a short-cup gizmo to convert quickly a multichannel layer into a deep image stream by normalising the 'distance form camera' channel output by your 3D render software.
Offset Deep data