Robustly separating sound components in human body based on 2-ch ICA and EM algorithm with dirichlet distribution.
Kenji HashiodaniS. TakadaYohei FukumizuHironori YamauchiYoshimasa KurumiTohru TaniPublished in: BHI (2012)
Keyphrases
- human body
- em algorithm
- mixture model
- dirichlet distribution
- expectation maximization
- parameter estimation
- maximum likelihood
- gaussian mixture model
- body parts
- hyperparameters
- generative model
- probabilistic model
- pose estimation
- three dimensional
- maximum a posteriori
- gaussian mixture
- unsupervised learning
- discrete data
- bayesian framework
- probability density function
- gaussian distribution
- incomplete data
- face recognition
- principal component analysis
- image sequences