I am trying to understand the PointNet network for dealing with point clouds and struggling with understanding the difference between FC and MLP:
"FC is fully connected layer operating on each point. MLP is multi-layer perceptron on each point."
I understand how fully connected layers are used to classify and I previously thought, was that MLP was the same thing but it seems varying academic papers have a differing definition from each other and from general online courses. In PointNet they seem to be used to mean different things?
My understanding was that up until the max pool in the network all the layer are run independently for each other for each of the N points and then the outputs of the nx1024 layer are max pooled. Why is it described a shared MLP then?
How does an MLP with an input 3 coordinates lead to an output from the MLP of 64? Or from 64 to 1024 even?
If this MLP is actually taking inputs from all of the N points, how do 1024 points lead to 64 wide MLP layer?
What does it mean by global and local features?