Data Fusion Applied to Feature Based Stereo Algorithms
In this chapter we describe a theoretical formulation for stereo (this was first proposed by Yuille, Geiger and Bülthoff in ) in terms of the Bayesian approach to vision outlined in chapters 2 and 3, in particular in terms of coupled Markov Random Fields. We show that this formalism is rich enough to contain most of the elements used in standard stereo theories.
KeywordsPartition Function Data Fusion Level Theory Epipolar Line Matching Element
Unable to display preview. Download preview PDF.