By Topic

Detect and track the dynamic deformation human body with the active shape model modified by motion vectors

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
Jia Ma ; Fac. of Eng., Univ. of Tokushima, Tokushima, Japan ; Ren, F.

At present, many systems are being used in the detection of human body from video, especially the moving human body. These systems usually recognize the human body features, such as learning the human body features by training, recognition of the skin color or body shape, and identify the specifically movement of human body. These methods have some problems. The training methods require a lot of training data, but only identify specific targets. The methods of identify the shape of human body and skin color are a great influenced by the video quality and content. The methods of detecting motion features are often complex and greatly relied to the 3D data. In this study, we propose an Active Shape Model (ASM) modified by the motion vectors to detect and track the deformed human body for action. Active shape model is used to identify the particular shapes; it is also a training method. We use kinds of original human body model modified by motion vectors, instead of training result. Thus the active shape model can detect the human body after deformation. So that human body won't be lost or repeated detected. In this method, we use Optical-flow method to get the motion vectors. The function of the method is conformed by experiment and the detecting success rate of moving human body is 94%. The experimental results and discussion are also presented.

Published in:

Cloud Computing and Intelligence Systems (CCIS), 2011 IEEE International Conference on

Date of Conference:

15-17 Sept. 2011