Self-tuning density estimation based on Bayesian averaging of adaptive kernel density estimations yields state-of-the-art performance.
Christofer L. BäcklinClaes R. AnderssonMats G. GustafssonPublished in: Pattern Recognit. (2018)
Keyphrases
- density estimation
- kernel density
- mixture model
- mixture modeling
- mean shift
- outlier detection
- probability density
- probability density function
- density function
- gaussian mixture model
- parzen window
- hidden markov models
- multivariate gaussian distribution
- kernel density estimation
- nonparametric density estimation
- maximum likelihood
- bayesian networks
- training data
- density estimators
- em algorithm
- density estimates
- computer vision
- learning algorithm
- data sets