Computer Vision Low Level API for Python/ZMQ access

It was mentioned that the CV has a low level and high level api. I can see that we can access the CV stuff via the high level nodeJS app api, but I was hoping to interface with it using python and zmq via malos. I see that the protobuffs have the config stuff for it, but in trying to reverse engineer the high level examples I couldn’t exactly find how the CV modules were talking to the camera hardware or the lower level malos-eye api. Is there any documentation on this?


I’m interested in this too.

A little late but this might be helpful: