Bayesian depth map interpolation using edge driven Markov Random Fields.
Stefania ColonneseStefano RinauroGaetano ScaranoPublished in: CompIMAGE (2012)
Keyphrases
- markov random field
- depth map
- image prior
- depth information
- multi view
- stereo matching
- belief propagation
- three dimensional
- d scene
- image restoration
- parameter estimation
- mrf model
- graph cuts
- maximum a posteriori
- high quality
- higher order
- energy function
- image sequences
- image segmentation
- depth estimation
- input image
- object boundaries
- high resolution
- energy minimization
- random fields
- potential functions
- gauss markov random fields
- edge information
- maximum likelihood
- super resolution
- edge detection
- bayesian networks
- low resolution
- pairwise
- posterior distribution
- depth images
- computer vision
- input data
- feature vectors
- viewpoint
- neighboring pixels
- image processing
- super resolved