Efficient Belief propagation for Early Vision课件.pptVIP

Efficient Belief propagation for Early Vision;perface; Markov random field (MRF) models Message passing 3 techniques to reduce the time 1.Computing Messages update in linear time 2.BP on the bipartite graph message passing schedule 3.Multi-Grid BP ;MRF models; the cost of assigning labels and to two neighboring pixels (discontinuity cost) , generally based on the difference between labels in low-level vision problem so we use the cost of assigning label to pixel (data cost ) ;Message passing;In general? . Ex 1.; Rules: Or: Ex 2. (complex) ;Message passing; be the message that node p sends to a neighboring node q at iteration t are initialized to 0 a vector of n_labels dimension Step 1. Compute messages:;;*;*;Step 3. minimizes individually at each node is selected time complexity: n number of pixels k number of possible labels ( n_lables ) T number of iterations;3 techniques;1.Computing Messages;Potts model ;truncated linear model First. Consider the pure linear model message compute ;lower envelope ;two-pass algorithm;two-pass algorithm;second. Use Potts model to computation;Potts model ;2.BP on the Grid Graph;Update ;So we alternate update the messages from A and from B scheme: when is odd , update when is even , update if is odd(even) then;Moreover , store new messages in the old one’s memory space(independent) The new messages are nearly the same as the messages in standard scheme when BP converges , messages converges to the same fixed point Turns to;3.Multi-Grid BP;Details: 0-th level: the original labeling problem i-th level: blocks of pixels grouped together . labelings where all the pixels in a block are assigned the same label level i get estim