Recognition of human activities using a multiclass relevance vector machine

He, Weihua; Yow, Kin Choong; Guo, Yongcai

doi:doi:10.1117/1.OE.51.1.017202

GENERAL INFORMATION

You are not logged in Logged Out Log In

Optical Engineering / Volume 51 / Issue 1 / Machine Vision, Pattern Recognition

Previous Article | Next Article

Recognition of human activities using a multiclass relevance vector machine

Opt. Eng. 51, 017202 (Feb 06, 2012); http://dx.doi.org/10.1117/1.OE.51.1.017202

Alerts
- Alert Me When Cited
- Alert Me When Corrected
Tools
Share
- Email Abstract
- Connotea
- CiteULike
- del.icio.us
- BibSonomy
- Tweet this Article
- Add to Facebook

Abstract

References (31)

Citing Articles (6)

Weihua He and Yongcai Guo

Chongqing University, Key Laboratory of Optoelectronic Technology and Systems, Ministry of Education, Chongqing 400030, China

Kin Choong Yow

Nanyang Technological University, School of Computer Engineering, N4-B2c-06, Nanyang Avenue, Singapore 639798, Singapore

We address the issue of human activity recognition by introducing the multiclass relevance vector machine (mRVM), the current state-of-the-art kernel machine learning technology given the multiclass classification problems (actually, activity recognition can commonly be viewed as a multiclass classification problem). Under our proposed recognition framework, the required procedure consists of three functional cascade modules: a. detecting the human silhouette blobs from the image sequence by the background subtraction method; b. extracting the shape and the motion features from the variation energy image (VEI); and c. sending the obtained features to the mRVM and recognizing the human activity. There are two types of mRVM: the constructive mRVM₁ and the top-down mRVM₂. We performed 10 times three-fold cross-validation on the Weizmann benchmark data set to examine the effectiveness of the proposed method. We also compared our method with other existing approaches, and the experimental results show that the proposed method offers superior performance. In summary, the mRVM, especially the mRVM₂, has advantages both in terms of recognition rate and sparsity, along with a simple feature extraction process. The mRVM also significantly simplifies the classification process, by comparison with traditional binary-tree style multiclass classifiers.

History

Received May 06, 2011
Accepted Nov 08, 2011
Revised Oct 02, 2011
Published online Feb 06, 2012

Digital Object Identifier

http://dx.doi.org/10.1117/1.OE.51.1.017202

Citation

Weihua He, Kin Choong Yow and Yongcai Guo, "Recognition of human activities using a multiclass relevance vector machine", Opt. Eng. 51, 017202 (Feb 06, 2012); http://dx.doi.org/10.1117/1.OE.51.1.017202