SGDNet: An End-to-End Saliency-Guided Deep Neural Network for No-Reference Image Quality Assessment.
Sheng YangQiuping JiangWeisi LinYongtao WangPublished in: ACM Multimedia (2019)
Keyphrases
- end to end
- image quality assessment
- neural network
- structural similarity
- quality assessment
- natural scene statistics
- image quality
- image quality measures
- natural scenes
- human visual system
- reduced reference
- image database
- quality metrics
- correlation coefficient
- natural images
- saliency map
- neural network model
- visual attention
- quality evaluation
- human perception
- subjective quality
- distorted images
- visual perception
- structural similarity index
- quality measures
- structural information
- perceptual image quality
- image quality metrics
- kullback leibler divergence
- data warehouse
- reconstructed image
- high resolution
- object recognition
- high quality
- computer vision