Unsupervised Motion Representation Learning with Capsule Autoencoders
This addresses the problem of learning robust motion representations without supervision for applications like human action recognition, though it appears incremental as it builds on capsule autoencoders.
The paper tackles the challenge of unsupervised motion representation learning by proposing the Motion Capsule Autoencoder (MCAE), which models motion in a two-level hierarchy with transformation-invariant templates, achieving state-of-the-art performance on skeleton-based action recognition and better results on a new dataset with fewer parameters.
We propose the Motion Capsule Autoencoder (MCAE), which addresses a key challenge in the unsupervised learning of motion representations: transformation invariance. MCAE models motion in a two-level hierarchy. 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. MCAE is evaluated on a novel Trajectory20 motion dataset and various real-world skeleton-based human action datasets. Notably, it achieves better results than baselines on Trajectory20 with considerably fewer parameters and state-of-the-art performance on the unsupervised skeleton-based action recognition task.