Package il2.inf.bp

Class Summary
BeliefPropagation some implementation notes: = we perform message-passing on the factor graph representation of a given Table[] array.
MaxProduct  
MaxProductLinearSmoothness Max-Product BP with more efficient message updates as in "Efficient Belief Propagation for Early Vision" by Felzenszwalb and Huttenlocher CVPR 2004, IJCV 2006 this assumes smoothness cost: \rho_p(d_s,d_t) = min(\lambda |d_s-d_t|,T) as in "Symmetric Stereo Matching for Occlusion Handling" by Sun, Li, Kang, and Shum CVPR 2005
 



Copyright 2010 UCLA Automated Reasoning Group