dipIQ: Blind Image Quality Assessment by Learning-to-Rank Discriminable Image Pairs.
Kede MaWentao LiuTongliang LiuZhou WangDacheng TaoPublished in: CoRR (2019)
Keyphrases
- image pairs
- learning to rank
- blind image quality assessment
- single image
- natural scene statistics
- ranking functions
- image quality
- image registration
- information retrieval
- image matching
- loss function
- ranking svm
- evaluation measures
- input image
- image quality assessment
- evaluation metrics
- document retrieval
- natural images
- retrieval systems
- visual quality
- quality assessment
- test collection
- collaborative filtering
- data mining
- object recognition
- feature space
- supervised learning
- matching algorithm