Services on Demand
Journal
Article
Indicators
Cited by SciELO
Access statistics
Related links
Cited by Google
Similars in SciELO
Similars in Google
Share
TecnoLógicas
Print version ISSN 0123-7799On-line version ISSN 2256-5337
Abstract
CABEZAS, Holman S. and SARMIENTO, Wilson J.. Evaluation of Models for Gesture Recognition from Biometric Signals of a Person with Reduced Mobility. TecnoL. [online]. 2019, vol.22, n.spe, pp.34-48. ISSN 0123-7799. https://doi.org/10.22430/22565337.1513.
This paper compares the results of three computational models (pattern recognition, hidden Markov models, and bag of features) for recognizing the hand gestures of a user with reduced mobility using biometric signal processing. The evaluation of the models included eight gestures co-designed with a person with reduced mobility. The models were evaluated using a cross-validation scheme, calculating sensitivity and precision metrics, and a data set of ten repetitions of each gesture. It can be concluded that the bag-of-features model achieved the best performance considering the two metrics under evaluation; the traditional pattern recognition model, using vector support machines, produced the most stable results; and the hidden Markov models had the lowest performance.
Keywords : Gesture recognition; Human computer interaction; Signal processing; Machine learning; Pattern recognition.