Action recognition using local joints structure and histograms of 3D joints

Yan Liang, Wanxuan Lu, Wei Liang, Yucheng Wang

科研成果: 书/报告/会议事项章节会议稿件同行评审

8 引用 (Scopus)

摘要

In this paper, we present a method for human action recognition using local joints structure and histograms of 3D joints. Global features like histograms of 3D joints [12] ignore the local structure information of the human body joints, which is also essential for accurate action recognition. To address this problem, we propose a local joints structure feature as a complement, and combine both global and local features for posture description in our method. Then, linear discriminant analysis is used to reduce the feature dimension, and k-means clustering is utilized to generate codewords. Finally, these codewords are treated as discrete symbols for training hidden Markov models (HMMs) which are used for action recognition. Experimental results demonstrate that our method has better performance than other methods when testing on UTKinect-Action Dataset and MSR Action3D dataset.

源语言英语
主期刊名Proceedings - 2014 10th International Conference on Computational Intelligence and Security, CIS 2014
出版商Institute of Electrical and Electronics Engineers Inc.
185-188
页数4
ISBN(电子版)9781479974344
DOI
出版状态已出版 - 20 1月 2015
活动10th International Conference on Computational Intelligence and Security, CIS 2014 - Kunming, Yunnan, 中国
期限: 15 11月 201416 11月 2014

出版系列

姓名Proceedings - 2014 10th International Conference on Computational Intelligence and Security, CIS 2014

会议

会议10th International Conference on Computational Intelligence and Security, CIS 2014
国家/地区中国
Kunming, Yunnan
时期15/11/1416/11/14

指纹

探究 'Action recognition using local joints structure and histograms of 3D joints' 的科研主题。它们共同构成独一无二的指纹。

引用此