Missing texture reconstruction via power spectrum-based sparse representation.
Yuma TanakaTakahiro OgawaMiki HaseyamaPublished in: GCCE (2015)
Keyphrases
- power spectrum
- sparse representation
- compressed sensing
- reconstruction error
- compressive sensing
- bank of gabor filters
- natural images
- sparse reconstruction
- frequency domain
- sparse coding
- dictionary learning
- face recognition
- random projections
- sparsity constraints
- object tracking
- texture features
- signal processing
- texture images
- higher order
- texture analysis
- image patches
- image classification
- high dimensional data
- object recognition
- image representation
- image reconstruction
- missing data
- joint optimization
- dimensionality reduction
- computer vision
- data mining
- high resolution
- image processing
- multiscale
- denoising
- input image
- probabilistic model