RGB-D based daily activity recognition for service robots using clustering with Gaussian Mixtures and FastDTW.
Kun QianGe GaoFang FangLiangjun ZhangPublished in: ICARM (2016)
Keyphrases
- gaussian mixture
- service robots
- daily activities
- activity recognition
- microsoft kinect
- clustering algorithm
- expectation maximization
- unsupervised learning
- depth information
- home environment
- k means
- closed form
- em algorithm
- home environments
- density estimation
- human activities
- object recognition
- human robot interaction
- gaussian mixture model
- probability density function
- mixture model
- covariance matrix
- human body
- depth data
- outlier detection
- data points
- bayesian networks
- feature extraction
- action recognition
- human computer interaction
- hidden markov models