Maximum likelihood for blind separation and deconvolution of noisy signals using mixture models.
Eric MoulinesJean-François CardosoElisabeth GassiatPublished in: ICASSP (1997)
Keyphrases
- blind separation
- mixture model
- maximum likelihood
- em algorithm
- blind source separation
- gaussian mixture model
- expectation maximization
- source signals
- independent component analysis
- blind deconvolution
- additive noise
- gaussian distribution
- hyperparameters
- denoising
- point spread function
- motion blur
- generative model
- blurred images
- maximum a posteriori
- image restoration
- source images
- unsupervised learning