A maximum a posteriori estimate for the source separation problem with statistical knowledge about the mixing matrix.
Jorge IgualAndrés CamachoPablo BernabeuLuis VergaraPublished in: Pattern Recognit. Lett. (2003)
Keyphrases
- source separation
- maximum a posteriori
- mixing matrix
- blind source separation
- independent component analysis
- markov random field
- map estimation
- maximum likelihood
- image reconstruction
- denoising
- em algorithm
- temporal structure
- expectation maximization
- random projections
- computer vision
- independent components
- single channel
- hyperparameters
- energy function
- prior knowledge
- feature vectors
- gaussian process
- blind separation
- face recognition
- hyperspectral images
- speech signal
- bayesian framework
- non stationary
- signal processing