Capsule Networks

An alternative to pooling in convolutional neural networks.

Problems with Pooling

Sometimes when outputting a smaller representation of an image, spatial information is lost. We can use capsule networks to avoid this.

What are Capsule Networks?

They detect parts in an object within the context of the spatial information.

Capsule networks are made of parent and child nodes that build up a complete picture of an object.

What are Capsules?

Capsules are a collection of nodes. Each node contains information about a specific part (width, orientation, colour), and outputs a vector with:

![Cat face, recognized in different orientations using magnitude and orientation](/Users/lauradang/Programming_Notes/Machine Learning/Convolutional Neural Networks/capsule-cat.png)

Pytorch Implementation

Here's a Github repo of a PyTorch implementation.

PreviousConvolutional Neural Networks NextCNN Input Output Shape

Last updated 3 years ago

Was this helpful?

What are Capsules?

Capsules are a collection of nodes. Each node contains information about a specific part (width, orientation, colour), and outputs a vector with:

Magnitude ( $m$ ) = the probability that a part exists; a value between 0 and 1.

Orientation ( $\theta$ ) = the state of the part properties

![Cat face, recognized in different orientations using magnitude and orientation](/Users/lauradang/Programming_Notes/Machine Learning/Convolutional Neural Networks/capsule-cat.png)