the Motivation of Capsules:
Dynamic Routing Between Capsules
•
Receptive field is generally understood by pixel level operations such
as 3×3 or 5×5, so the convolutional layer always tries to understand
local features and information. CNN loses coordinate frame because of
pooling.
•
A capsule is a group of neurons whose activity vector represents the
instantiation parameters of a specific type of entity.
•
Using the length of the activity vector to represent the probability that
the entity exists and its orientation to represent the instantiation
parameters.