Sanghyeok Oh and Keechul Jung
ViewPoint Insensitive Human Pose Recognition using Neural Network and CUDA
2911 - 2914
2009
3
12
International Journal of Computer and Information Engineering
https://publications.waset.org/pdf/5347
https://publications.waset.org/vol/36
World Academy of Science, Engineering and Technology
Although lots of research work has been done for
human pose recognition, the viewpoint of cameras is still critical
problem of overall recognition system. In this paper, viewpoint
insensitive human pose recognition is proposed. The aims of the
proposed system are viewpoint insensitivity and realtime processing.
Recognition system consists of feature extraction module, neural
network and realtime feed forward calculation. First, histogrambased
method is used to extract feature from silhouette image and it is
suitable for represent the shape of human pose. To reduce the
dimension of feature vector, Principle Component Analysis(PCA) is
used. Second, realtime processing is implemented by using Compute
Unified Device Architecture(CUDA) and this architecture improves
the speed of feedforward calculation of neural network. We
demonstrate the effectiveness of our approach with experiments on
real environment.
Open Science Index 36, 2009