CT-SRCNN: Cascade Trained and Trimmed Deep Convolutional Neural Networks for Image Super Resolution.
Haoyu RenMostafa El-KhamyJungwon LeePublished in: WACV (2018)
Keyphrases
- convolutional network
- convolutional neural networks
- image super resolution
- low resolution
- super resolution
- dictionary learning
- image registration
- high resolution
- kernel regression
- sparse representation
- maximum a posteriori
- edge preserving
- image patches
- low resolution images
- map estimation
- three dimensional
- data sets
- image classification
- markov random field
- principal component analysis