Action recognition using ensemble weighted multi-instance learning

Guang Chen, Manuel Giuliani, Daniel Clarke, Andre Gaschler, Alois C. Knoll

Abstract

This paper deals with recognizing human actions in depth video data. Current state-of-the-art action recognition methods use hand-designed features, which are difficult to produce and time-consuming to extend to new modalities. In this paper, we propose a novel, 3.5D representation of a depth video for action recognition. A 3.5D graph of the depth video consists of a set of nodes that are the joints of the human body. Each joint is represented by a set of spatio-temporal features, which are computed by an unsupervised learning approach. However, if occlusions occur, the 3D positions of the joints are noisy which increases the intra-class variations in action classes. To address this problem, we propose the Ensemble Weighted Multi-Instance Learning approach (EnwMi) for the action recognition task. It considers the class imbalance and intra-class variations. We formulate the action recognition task with depth videos as a weighted multi-instance problem. We further integrate an ensemble learning method into the weighted multi-instance learning framework. Our approach is evaluated on Microsoft Research Action3D dataset, and the results show that it outperforms state-of-the-art methods.
BibTeX
@inproceedings{chen2014action,
  author = {Guang Chen and Manuel Giuliani and Daniel Clarke and Andre Gaschler and Alois C. Knoll},
  title = {{Action recognition using ensemble weighted multi-instance learning}},
  booktitle = {IEEE International Conference on Robotics and Automation},
  year = {2014},
  doi = {10.1109/icra.2014.6907519},
}

Loading related content...

Figure