VCISR: Blind Single Image Super-Resolution with Video Compression Synthetic Data.
Boyang WangBowen LiuShiyu LiuFengyu YangPublished in: CoRR (2023)
Keyphrases
- synthetic data
- video compression
- single image super resolution
- super resolution
- motion estimation
- sparse coding
- motion compensation
- low resolution
- video coding
- dictionary learning
- motion compensated
- image patches
- video data
- real image data
- data sets
- high resolution
- motion vectors
- compressive sensing
- sparse representation
- real world
- compression ratio
- natural images
- image classification
- optical flow
- image processing
- database
- image prior
- prediction error
- image sequences
- signal processing
- motion field
- computer vision
- machine learning
- video sequences
- test images
- moving objects
- unsupervised learning