Discriminative Key Pose Extraction Using Extended LC-KSVD for Action Recognition
Authors:
- Lijuan ZhouLZLijuan Zhou
- Wanqing LiWLWanqing Li
- Yuyao ZhangYZYuyao Zhang
- Philip OgunbonaPOPhilip Ogunbona
- Duc Thanh NguyenDNDuc Thanh Nguyen
- Hanling ZhangHZHanling Zhang
- 1 November 2014
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
Abstract
This paper presents a method for extracting discriminative key poses for skeleton-based action recognition. Poses are represented by normalized joint locations, velocities and accelerations of skeleton joints. An extended label consistent K-SVD (ELC-KSVD) algorithm is proposed for learning the common and action-specific dictionaries. Discriminative key poses are represented by the atoms of the action-specific dictionaries. With the specific dictionaries, sparse codes are obtained for representing action instances through max pooling and temporal pyramid. A SVM classifier is trained for action recognition. The proposed method was evaluated on the MSRC-12 gesture and MSR-Action 3D datasets. Experimental results have shown that the proposed method is effective in extracting discriminative key poses.Keywords
This publication has 27 references indexed in Scilit:
- Silhouette-based human action recognition using sequences of key posesPattern Recognition Letters, 2013
- Learning Discriminative Key Poses for Action RecognitionIEEE Transactions on Cybernetics, 2013
- Action recognition by learning discriminative key poses2011
- Selecting Key Poses on Manifold for Pairwise Action RecognitionIEEE Transactions on Industrial Informatics, 2011
- Learning a discriminative dictionary for sparse coding via label consistent K-SVD2011
- Action Recognition Using Sparse Representation on Covariance Manifolds of Optical Flow2010
- Recognizing Human Actions Using Key Poses2010
- Discriminative K-SVD for dictionary learning in face recognition2010
- Action recognition using exemplar-based embedding2008
- $rm K$-SVD: An Algorithm for Designing Overcomplete Dictionaries for Sparse RepresentationIEEE Transactions on Signal Processing, 2006