A probabilistic pairwise-preference predictor for image quality.
Amy R. ReibmanKenneth ShirleyChao TianPublished in: ICIP (2013)
Keyphrases
- image quality
- pairwise
- visual quality
- image data
- quality assessment
- pairwise comparison
- reconstructed image
- markov random field
- image resolution
- bit rate
- human visual system
- high quality
- compression ratio
- probabilistic model
- neural network
- loss function
- user preferences
- generative model
- bayesian networks
- high image quality
- pairwise comparisons
- higher order
- multi class
- standard test images
- similarity measure
- lossy compression
- preference relations
- spectral clustering
- image quality assessment
- quality metrics
- pixel values
- belief networks
- belief propagation
- human visual sensitivity