DeepCaps: Going Deeper with Capsule Networks

Abstract

Capsule Network is a promising concept in deep learning, yet its true potential is not fully realized thus far, providing sub-par performance on several key benchmark datasets with complex data. Drawing intuition from the success achieved by Convolutional Neural Networks (CNNs) by going deeper, we introduce DeepCaps, a deep capsule network architecture which uses a novel 3D convolution based dynamic routing algorithm. With DeepCaps, we surpass the state-of-the-art capsule domain networks results on CIFAR10, SVHN and Fashion MNIST, while achieving a 68% reduction in the number of parameters. Further, we propose a class independent decoder network, which strengthens the use of reconstruction loss as a regularization term. This leads to an interesting property of the decoder, which allows us to identify and control the physical attributes of the images represented by the instantiation parameters.

Cite

Text

Rajasegaran et al. "DeepCaps: Going Deeper with Capsule Networks." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019. doi:10.1109/CVPR.2019.01098

Markdown

[Rajasegaran et al. "DeepCaps: Going Deeper with Capsule Networks." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019.](https://mlanthology.org/cvpr/2019/rajasegaran2019cvpr-deepcaps/) doi:10.1109/CVPR.2019.01098

BibTeX

@inproceedings{rajasegaran2019cvpr-deepcaps,
  title     = {{DeepCaps: Going Deeper with Capsule Networks}},
  author    = {Rajasegaran, Jathushan and Jayasundara, Vinoj and Jayasekara, Sandaru and Jayasekara, Hirunima and Seneviratne, Suranga and Rodrigo, Ranga},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2019},
  doi       = {10.1109/CVPR.2019.01098},
  url       = {https://mlanthology.org/cvpr/2019/rajasegaran2019cvpr-deepcaps/}
}