Research Article
Real time gesture recognition using continuous time recurrent neural networks
@INPROCEEDINGS{10.4108/bodynets.2007.149, author={Gonzalo Bailador and Daniel Roggen and Gerhard Tr\o{}ster and Graci\^{a}n Trivi\`{o}o}, title={Real time gesture recognition using continuous time recurrent neural networks}, proceedings={2nd International ICST Conference on Body Area Networks}, publisher={ICST}, proceedings_a={BODYNETS}, year={2007}, month={6}, keywords={Gesture Recognition Recurrent Neural Networks}, doi={10.4108/bodynets.2007.149} }- Gonzalo Bailador
Daniel Roggen
Gerhard Tröster
Gracián Triviño
Year: 2007
Real time gesture recognition using continuous time recurrent neural networks
BODYNETS
ICST
DOI: 10.4108/bodynets.2007.149
Abstract
This paper presents a new approach to the problem of gesture recognition in real time using inexpensive accelerometers. This approach is based on the idea of creating specialized signal predictors for each gesture class. These signal predictors forecast future acceleration values from current ones. The errors between the measured acceleration of a given gesture and the predictors are used for classification. This approach is modular and allows for seamless inclusion of new gesture classes. These predictors are implemented using Continuous Time Recurrent Neural Networks (CTRNN). On the one hand, this kind of networks exhibits rich dynamical behaviour that is useful in gesture recognition and on the other, they have a relatively low computational cost that is interesting feature for real time systems.
