TY - GEN
T1 - (BP)2
T2 - 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR
AU - Nwogu, Ifeoma
AU - Corso, Jason J.
PY - 2008
Y1 - 2008
N2 - Belief Propagation (BP) can be very useful and efficient for performing approximate inference on graphs. But when the graph is very highly connected with strong conflicting interactions, BP tends to fail to converge. Generalized Belief Propagation (GBP) provides more accurate solutions on such graphs, by approximating Kikuchi free energies, but the clusters required for the Kikuchi approximations are hard to generate. We propose a new algorithmic way of generating such clusters from a graph without exponentially increasing the size of the graph during triangulation. In order to perform the statistical region labeling, we introduce the use of superpixels for the nodes of the graph, as it is a more natural representation of an image than the pixel grid. This results in a smaller but much more highly interconnected graph where BP consistently fails. We demonstrate how our version of the GBP algorithm outperforms BP on synthetic and natural images and in both cases, GBP converges after only a few iterations.
AB - Belief Propagation (BP) can be very useful and efficient for performing approximate inference on graphs. But when the graph is very highly connected with strong conflicting interactions, BP tends to fail to converge. Generalized Belief Propagation (GBP) provides more accurate solutions on such graphs, by approximating Kikuchi free energies, but the clusters required for the Kikuchi approximations are hard to generate. We propose a new algorithmic way of generating such clusters from a graph without exponentially increasing the size of the graph during triangulation. In order to perform the statistical region labeling, we introduce the use of superpixels for the nodes of the graph, as it is a more natural representation of an image than the pixel grid. This results in a smaller but much more highly interconnected graph where BP consistently fails. We demonstrate how our version of the GBP algorithm outperforms BP on synthetic and natural images and in both cases, GBP converges after only a few iterations.
UR - http://www.scopus.com/inward/record.url?scp=51949101844&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=51949101844&partnerID=8YFLogxK
U2 - 10.1109/CVPR.2008.4587371
DO - 10.1109/CVPR.2008.4587371
M3 - Conference contribution
AN - SCOPUS:51949101844
SN - 9781424422432
T3 - 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR
BT - 26th IEEE Conference on Computer Vision and Pattern Recognition, CVPR
Y2 - 23 June 2008 through 28 June 2008
ER -