Overcomplete Dictionary Learning for Nonnegative Sparse Representation with an ℓ_p-Norm Constraint Based on Majorize-Minimization.
Shiori IshikuroShuxue DingXiang LiPublished in: SMC (2018)
Keyphrases
- sparse representation
- dictionary learning
- nonnegative matrix factorization
- negative matrix factorization
- norm minimization
- sparse coding
- objective function
- graph embedding
- image super resolution
- image classification
- face recognition
- orthogonal matching pursuit
- group sparsity
- compressive sensing
- sparse codes
- high dimensional data
- random projections
- image patches
- sparsity constraints
- image representation
- dimensionality reduction
- compressed sensing
- joint optimization
- data sets
- single image super resolution
- principal component analysis
- signal processing
- regularization term
- matrix factorization
- image features
- small number
- discriminative power
- probabilistic model
- test images