Skip to yearly menu bar Skip to main content


Poster

Unsupervised Learning of View-invariant Action Representations

Junnan Li · Yongkang Wong · Qi Zhao · Mohan Kankanhalli

Room 517 AB #122

Keywords: [ Unsupervised Learning ] [ Activity and Event Recognition ] [ Tracking and Motion in Video ]


Abstract:

The recent success in human action recognition with deep learning methods mostly adopt the supervised learning paradigm, which requires significant amount of manually labeled data to achieve good performance. However, label collection is an expensive and time-consuming process. In this work, we propose an unsupervised learning framework, which exploits unlabeled data to learn video representations. Different from previous works in video representation learning, our unsupervised learning task is to predict 3D motion in multiple target views using video representation from a source view. By learning to extrapolate cross-view motions, the representation can capture view-invariant motion dynamics which is discriminative for the action. In addition, we propose a view-adversarial training method to enhance learning of view-invariant features. We demonstrate the effectiveness of the learned representations for action recognition on multiple datasets.

Live content is unavailable. Log in and register to view live content