Forward Super-Resolution: How Can GANs Learn Hierarchical Generative Models for Real-World Distributions.
Zeyuan Allen-ZhuYuanzhi LiPublished in: ICLR (2023)
Keyphrases
- super resolution
- generative model
- low resolution
- high resolution
- image reconstruction
- probabilistic model
- low resolution images
- discriminative models
- mixture model
- high quality
- motion estimation
- conditional random fields
- data sets
- discriminative learning
- sparse coding
- bayesian framework
- em algorithm
- semi supervised
- prior knowledge
- depth map
- topic models
- image patches
- probability distribution
- markov random field
- semi supervised learning
- random variables
- feature vectors
- object recognition
- image segmentation
- image processing
- feature selection