Combining unsupervised learning and discrimination for 3D action recognition

Chen, Guang; Clarke, Daniel; Giuliani, Manuel; Gaschler, Andre; Knoll, Alois

Abstract

© 2014 Elsevier B.V. Previous work on 3D action recognition has focused on using hand-designed features, either from depth videos or 2D videos. In this work, we present an effective way to combine unsupervised feature learning with discriminative feature mining. Unsupervised feature learning allows us to extract spatio-temporal features from unlabeled video data. With this, we can avoid the cumbersome process of designing feature extraction by hand. We propose an ensemble approach using a discriminative learning algorithm, where each base learner is a discriminative multi-kernel-learning classifier, trained to learn an optimal combination of joint-based features. Our evaluation includes a comparison to state-of-the-art methods on the MSRAction 3D dataset, where our method, abbreviated EnMkl, outperforms earlier methods. Furthermore, we analyze the efficiency of our approach in a 3D action recognition system.

• Journal Article Type: Article
• Acceptance Date: Aug 16, 2014
• Online Publication Date: Aug 23, 2014
• Publication Date: 2015-05
• Journal: Signal Processing
• Print ISSN: 0165-1684
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 110
• Pages: 67-81
• DOI: https://doi.org/10.1016/j.sigpro.2014.08.024
• Keywords: human action recognition, depth camera, unsupervised learning, multi-kernel learning, ensemble learning
• Public URL: https://uwe-repository.worktribe.com/output/835010
• Publisher URL: http://dx.doi.org/10.1016/j.sigpro.2014.08.024
• Related Public URLs: http://www.sciencedirect.com/science/article/pii/S0165168414003880