Synthetic Training Data Generation for Activity Monitoring and Behavior Analysis

Authors:
Dorothy Monekosso;Paolo Remagnino
Affiliations:
CSRI, University of Ulster, Jordanstown, UK;CISM, Kingston University, London, UK
Venue:
AmI '09 Proceedings of the European Conference on Ambient Intelligence
Year:
2009

Citing 7
Cited 1

EasyLiving: Technologies for Intelligent Environments

HUC '00 Proceedings of the 2nd international symposium on Handheld and Ubiquitous Computing
Learning to Detect User Activity and Availability from a Variety of Sensor Data

PERCOM '04 Proceedings of the Second IEEE International Conference on Pervasive Computing and Communications (PerCom'04)
Resampling for Face Detection by Self-Adaptive Genetic Algorithm

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 3 - Volume 03
Synthetic data generation technique in Signer-independent sign language recognition

Pattern Recognition Letters
Resampling for face recognition

AVBPA'03 Proceedings of the 4th international conference on Audio- and video-based biometric person authentication
Detecting individual activities from video in a smart home

KES'07/WIRN'07 Proceedings of the 11th international conference, KES 2007 and XVII Italian workshop on neural networks conference on Knowledge-based intelligent information and engineering systems: Part I
A fuzzy embedded agent-based approach for realizing ambient intelligence in intelligent inhabited environments

IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans

Towards the detection of unusual temporal events during activities using HMMs

Proceedings of the 2012 ACM Conference on Ubiquitous Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes a data generator that produces synthetic data to simulate observations from an array of environment monitoring sensors. The overall goal of our work is to monitor the well-being of one occupant in a home. Sensors are embedded in a smart home to unobtrusively record environmental parameters. Based on the sensor observations, behavior analysis and modeling are performed. However behavior analysis and modeling require large data sets to be collected over long periods of time to achieve the level of accuracy expected. A data generator - was developed based on initial data i.e. data collected over periods lasting weeks to facilitate concurrent data collection and development of algorithms. The data generator is based on statistical inference techniques. Variation is introduced into the data using perturbation models.