Unsupervised Motion Representation Learning with Capsule Autoencoders

1 code implementation • NeurIPS 2021 • Ziwei Xu, Xudong Shen, Yongkang Wong, Mohan S Kankanhalli

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

Action Recognition Representation Learning +2