In this video, we introduce our activity recognition system and demonstrate it in a naturalistic environment. We recognise the activities 'Sitting', 'Standing', 'Walking', 'Running', 'Jumping', 'Falling' and 'Lying' with four different Bayesian classifiers: A Naive Bayes classifier and a Bayesian Network classifier trained on a data set of 16 persons, both in a static version and a dynamic version that is enhanced by a first-order Hidden Markov Model (HMM) and evaluated with a grid based filter.
Link to this comment:
All Comments (0)