MetaIQA: Deep Meta-learning for No-Reference Image Quality Assessment.
Hancheng ZhuLeida LiJinjian WuWeisheng DongGuangming ShiPublished in: CoRR (2020)
Keyphrases
- meta learning
- image quality assessment
- image quality
- quality assessment
- learning tasks
- natural scene statistics
- structural similarity
- inductive learning
- model selection
- machine learning algorithms
- image quality measures
- natural images
- image database
- reduced reference
- human visual system
- human perception
- subjective quality
- correlation coefficient
- decision trees
- quality metrics
- visual perception
- quality evaluation
- image quality metrics
- machine learning
- natural scenes
- feature selection
- metamodel
- perceptual image quality
- kullback leibler divergence
- data mining
- base classifiers
- databases
- prior knowledge
- structural information
- text categorization
- knowledge base