Tracking People using Blob Detection? (new to Max)

Hey all, I'm new to max, but amazed by it's functionality as well as the community involved here! I am looking to create a work that uses a camera to track blobs in the frame to create a musical scale from left to right by segmenting the grid. Then is it possible to use color of the people interacting to determine pitch of the note? Any help or documentation would be greatly appreciated. (I have the cv.Jit package already)