Human posture recognition with the stochastic cognitive RAM network

Authors:
Weng Kin Lai;Imran M. Khan;George G. Coghill
Affiliations:
School of Technology, TARC, Kuala Lumpur, Malaysia;Dept. of Electrical and Computer Engineering, IIUM, Kuala Lumpur, Malaysia;Department of Computer and Electrical Engineering, University of Auckland, New Zealand
Venue:
ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part V
Year:
2012

Citing 4
Cited 0

Adaptive pattern recognition using goal seeking neurons

Pattern Recognition Letters
Shape Analysis and Classification: Theory and Practice

Shape Analysis and Classification: Theory and Practice
Real-time recognition of activity using temporal templates

WACV '96 Proceedings of the 3rd IEEE Workshop on Applications of Computer Vision (WACV '96)
Posture Estimation in Visual Surveillance of Archaeological Sites

AVSS '03 Proceedings of the IEEE Conference on Advanced Video and Signal Based Surveillance

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper examines a weightless neural network (WNN) for human posture recognition. Like all earlier weightless neural network models, the Cognitive RAM Network (CogRAM) learns in one pass through the data and due to its simplicity it can be fabricated in hardware. While it has shown good performance in earlier studies, it still suffers from the common problem of network saturation especially when it comes to high dimensional and poorly separated data in the feature space. Hence, we proposed the Stochastic CogRAM which has shown significant improvements when tested on the challenging human postures recognition problem. We also present some comparisons of the experimental results obtained from the popular K-Means clustering algorithm. Future research is outlined at the end of the paper.