Uncertainty of Atmospheric Motion Vectors by Sampling Tempered Posterior Distributions.
Patrick HéasFrédéric CérouMathias RoussetPublished in: CoRR (2022)
Keyphrases
- motion vectors
- posterior distribution
- metropolis hastings
- markov chain monte carlo
- motion estimation
- motion compensation
- video sequences
- probability distribution
- latent variables
- motion field
- parameter estimation
- video coding
- compressed domain
- bit rate
- bayesian framework
- computational complexity
- sampling algorithm
- block matching
- video data
- hyperparameters
- posterior probability
- maximum a posteriori
- random sampling
- closed form
- sample size
- conditional probabilities
- generative model
- multimedia
- probabilistic model
- prior knowledge
- maximum likelihood
- model selection