Institutions | About Us | Help | Gaeilge
rian logo


Mark
Go Back
Classification of Human Poses using a Vision based Technique
Kelly, Dan; Olivo, Paolo; Markham, Charles; McDonald, John; Caulfield, Brian; Fitzgerald, Diarmuid
This paper presents work being carried out to estimate human pose using vision based methods. The data acquisition system uses an active marker technique synchronized with a three camera stereo vision system. The locations of the markers are then used to reconstruct a skeleton representation of the human pose. PCA and clustering techniques are used to classify the pose.
Keyword(s): stereo image processing; computer vision; image classification; image reconstruction; image representation; principal component analysis
Publication Date:
2007
Type: Book chapter
Peer-Reviewed: Yes
Institution: Maynooth University
Citation(s): Kelly, Dan and Olivo, Paolo and Markham, Charles and McDonald, John and Caulfield, Brian and Fitzgerald, Diarmuid (2007) Classification of Human Poses using a Vision based Technique. In: IMVIP 2007. International Machine Vision and Image Processing Conference, 2007. IEEE, p. 209. ISBN 0769528872
Publisher(s): IEEE
File Format(s): other
Related Link(s): http://mural.maynoothuniversity.ie/8348/1/JM-Classification-2007.pdf
First Indexed: 2020-04-02 06:30:49 Last Updated: 2020-04-02 06:30:49