… our capsule receives (u1, u2 and u3 in the diagram) come from 3 other capsules in the layer below. Lengths of these vectors encode probabilities that lower-level capsules detected their corresponding objects and directions of the vectors encode some internal state of the detected objects. Let us assume that lower level capsules detect eyes, mouth and nose respectively and out capsule de…
…mechanism is not very good, because max pooling loses valuable information and also does not encode relative spatial relationships between features. We should use capsules instead, because they will encapsulate all important information about the s…