Objects are composed of a set of geometrically organized parts. We introducean unsupervised capsule autoencoder ( SCAE), which explicitly uses geometric relationships between parts toreason about objects. Since these relationships do not depend on the viewpoint, our model is robust to viewpoint changes.

We propose the Motion Capsule Autoencoder (MCAE), which addresses a key challenge in the unsupervised learning of motion representations: transformation invariance.

We show in the table below the notations grouped by the modules. The values used in our implementation are shown if applicable. The necessity of a two-layer hierarchy is briefly discussed in Section 3.3. In short, it is difficult for a single-layer hierarchy to capture long-time dependencies and variations. This section describes an empirical study where we compare MCAE with its single-layer correspondence.

Neural Information Processing Systems http://nips.cc/

We propose the Motion Capsule Autoencoder (MCAE), which addresses a key challenge in the unsupervised learning of motion representations: transformation invariance. In the lower level, a spatio-temporal motion signal is divided into short, local, and semantic-agnostic snippets. In the higher level, the snippets are aggregated to form full-length semantic-aware segments. For both levels, we represent motion with a set of learned transformation invariant templates and the corresponding geometric transformations by using capsule autoencoders of a novel design. This leads to a robust and efficient encoding of viewpoint changes.

We propose the Motion Capsule Autoencoder (MCAE), which addresses a key challenge in the unsupervised learning of motion representations: transformation invariance. In the lower level, a spatio-temporal motion signal is divided into short, local, and semantic-agnostic snippets. In the higher level, the snippets are aggregated to form full-length semantic-aware segments. For both levels, we represent motion with a set of learned transformation invariant templates and the corresponding geometric transformations by using capsule autoencoders of a novel design. This leads to a robust and efficient encoding of viewpoint changes.

An object can be seen as a geometrically organized set of interrelated parts. A system that makes explicit use of these geometric relationships to recognize objects should be naturally robust to changes in viewpoint, because the intrinsic geometric relationships are viewpoint-invariant. We describe an unsupervised version of capsule networks, in which a neural encoder, which looks at all of the parts, is used to infer the presence and poses of object capsules. The encoder is trained by backpropagating through a decoder, which predicts the pose of each already discovered part using a mixture of pose predictions. The parts are discovered directly from an image, in a similar manner, by using a neural encoder, which infers parts and their affine transformations. The corresponding decoder models each image pixel as a mixture of predictions made by affine-transformed parts. We learn object- and their part-capsules on unlabeled data, and then cluster the vectors of presences of object capsules. When told the names of these clusters, we achieve state-of-the-art results for unsupervised classification on SVHN (55%) and near state-of-the-art on MNIST (98.5%).