Dissecting Arbitrary-scale Super-resolution Capability from Pre-trained Diffusion Generative Models.
Ruibin LiQihua ZhouSong GuoJie ZhangJingcai GuoXinyang JiangYifei ShenZhenhua HanPublished in: CoRR (2023)
Keyphrases
- super resolution
- generative model
- low resolution
- high resolution
- probabilistic model
- motion estimation
- single image
- high quality
- em algorithm
- prior knowledge
- multi frame
- depth map
- semi supervised
- text classification
- maximum likelihood
- active learning
- image patches
- data sets
- expectation maximization
- d objects
- decision trees
- video data
- sparse coding