Convolutional Fully-Connected Capsule Network (CFC-CapsNet): A Novel and Fast Capsule Network

TL;DR

This paper introduces CFC-CapsNet, a new capsule network architecture that improves speed, accuracy, and parameter efficiency over traditional CapsNet, making it more suitable for complex datasets and real-world applications.

Contribution

The paper proposes a novel CFC layer for capsule creation, resulting in fewer, more powerful capsules and enhanced performance over conventional CapsNet.

Findings

Achieves higher accuracy on CIFAR-10, SVHN, Fashion-MNIST

Faster training and inference times

Uses fewer parameters than traditional CapsNet

Abstract

A Capsule Network (CapsNet) is a relatively new classifier and one of the possible successors of Convolutional Neural Networks (CNNs). CapsNet maintains the spatial hierarchies between the features and outperforms CNNs at classifying images including overlapping categories. Even though CapsNet works well on small-scale datasets such as MNIST, it fails to achieve a similar level of performance on more complicated datasets and real applications. In addition, CapsNet is slow compared to CNNs when performing the same task and relies on a higher number of parameters. In this work, we introduce Convolutional Fully-Connected Capsule Network (CFC-CapsNet) to address the shortcomings of CapsNet by creating capsules using a different method. We introduce a new layer (CFC layer) as an alternative solution to creating capsules. CFC-CapsNet produces fewer, yet more powerful capsules resulting in higher network accuracy. Our experiments show that CFC-CapsNet achieves competitive accuracy, faster training and inference and uses less number of parameters on the CIFAR-10, SVHN and Fashion-MNIST datasets compared to conventional CapsNet.