PPFM: Image denoising in photon-counting CT using single-step posterior sampling Poisson flow generative models.
Dennis HeinStaffan HolminTimothy SzczykutowiczJonathan S. MaltzMats DanielssonGe WangMats PerssonPublished in: CoRR (2023)
Keyphrases
- image denoising
- generative model
- single step
- posterior probability
- photon counting
- bayesian framework
- probabilistic model
- denoising
- three dimensional
- total variation
- natural images
- image processing
- semi supervised
- wavelet coefficients
- posterior distribution
- prior knowledge
- wavelet domain
- em algorithm
- conditional random fields
- expectation maximization
- multispectral
- flow field
- object categories
- hidden variables
- markov networks
- bayesian inference
- latent variables
- probability distribution
- multiscale
- conditional probabilities
- image analysis
- image reconstruction
- bayesian networks
- super resolution