I'm working with the Azure Kinect camera and are trying to detect if a person is holding an object in the hand/hands. I have attached two pictures illustrating the input data:
] 
So I have all positions and rotation data for the body minus the hands. Furthermore, the depth map showed in the picture.
I'm looking for a solution there can detect the bottle I'm holding in my hands on the depth map. The bottle is an example and it is not important to detect what object it is, only if there is an object in the hand/hands. The solution should be able to be implemented in python.
In advance thanks for your help.