Neural Networks: A Comprehensive Foundation
Neural Networks: A Comprehensive Foundation
Machine Learning
Machine Learning
The WEKA data mining software: an update
ACM SIGKDD Explorations Newsletter
Person Re-identification Using Spatial Covariance Regions of Human Body Parts
AVSS '10 Proceedings of the 2010 7th IEEE International Conference on Advanced Video and Signal Based Surveillance
Some issues of biometrics: technology intelligence, progress and challenges
International Journal of Information Technology and Management
Using HMMs for Discriminating Mobile from Static Objects in a 3D Occupancy Grid
ICTAI '11 Proceedings of the 2011 IEEE 23rd International Conference on Tools with Artificial Intelligence
| Hi-index | 0.00 |
In this paper, we consider the viability of using Microsoft Kinect sensor to extract skeleton points from walking subjects and use these points for biometric identification. We do so by capturing several subjects using the sensor, calculating the length of several body parts inferred from the extracted points and training a model for later classification using these lengths and labels identifying the subjects as training examples. We consider the cases where one wants to discriminate each subject individually and where only recognizing a single subject is enough, showing that in both cases a Nearest Neighbor algorithm is able to achieve high accuracy when considering a relatively small group of subjects. However, our approach requires a moderately large number of training examples and we discuss the impact of such caveat in certain scenarios. Finally, we consider the contribution of different combinations of body parts to the identification process.