how do I provide a depth data array, already salavta, to the scheleton tracker ??
Frustratingly, this is NOT possible with the nuitrack SDK. There is currently no way to pass an external depth data array to the user/ body / skeleton / hand or identity tracker - THOUGH it is something that has been requested previously by a number of people.
The current nuitrack runtime is bound and licenced to specific hardware devices - and it can only receive frame data directly from those devices. It cannot currently receive depth or RGB data from a file or from any other external source.
Is any processing of the image possible before feeding it to the tracker? I would at least like to be able to rotate it.
While it is not possible to directly manipulate the depth data before nuitrack gets it - for the most part
There is a rotation setting on the nuitrack.config - that can also be set at runtime.
“RotateAngle”: “0”, (0, 90 ,180, 270) are allowed values
There are also some filtering settings - if you are using a intel d435 sensor - but these are not that well documented and more relate to depth data cleanup.