A fast, massively parallel solver for large, irregular pairwise Markov random fields.
Daniel ThuerckMichael WaechterSven WidmerMax von BülowPatrick SeemannMarc E. PfetschMichael GoeselePublished in: High Performance Graphics (2016)
Keyphrases
- markov random field
- massively parallel
- pairwise
- higher order
- belief propagation
- fine grained
- graph cuts
- mrf model
- image restoration
- random fields
- parameter estimation
- parallel computing
- energy function
- maximum a posteriori
- conditional random fields
- map inference
- energy minimization
- high order
- map estimation
- potential functions
- loopy belief propagation
- potts model
- textured images
- iterative conditional
- message passing
- low level vision
- discriminative random fields
- efficient inference
- partition function
- image segmentation
- multi label
- optimal solution