Timezone: »

Unsupervised Motion Representation Learning with Capsule Autoencoders
Ziwei Xu · Xudong Shen · Yongkang Wong · Mohan Kankanhalli

Thu Dec 09 08:30 AM -- 10:00 AM (PST) @

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.

Author Information

Ziwei Xu (National University of Singapore)
Xudong Shen (National University of Singapore)
Yongkang Wong (National University of Singapore)
Mohan Kankanhalli (National University of Singapore,)

More from the Same Authors