Quaternion Capsule Networks
This work addresses the challenge of object recognition in unseen viewpoints for computer vision, though it is incremental as it builds on existing capsule network methods.
The paper tackled the problem of improving viewpoint generalization in capsule networks by representing pose information with quaternions, resulting in better generalization to novel viewpoints with fewer parameters and achieving on-par or better performance compared to state-of-the-art capsule architectures on benchmarking datasets.
Capsules are grouping of neurons that allow to represent sophisticated information of a visual entity such as pose and features. In the view of this property, Capsule Networks outperform CNNs in challenging tasks like object recognition in unseen viewpoints, and this is achieved by learning the transformations between the object and its parts with the help of high dimensional representation of pose information. In this paper, we present Quaternion Capsules (QCN) where pose information of capsules and their transformations are represented by quaternions. Quaternions are immune to the gimbal lock, have straightforward regularization of the rotation representation for capsules, and require less number of parameters than matrices. The experimental results show that QCNs generalize better to novel viewpoints with fewer parameters, and also achieve on-par or better performances with the state-of-the-art Capsule architectures on well-known benchmarking datasets.