Belief Propagation in Large, Highly Connected Graphs for 3D Part-Based Object Recognition.
Frank DiMaioJude W. ShavlikPublished in: ICDM (2006)
Keyphrases
- highly connected
- belief propagation
- object recognition
- graphical models
- message passing
- graph cuts
- markov random field
- stereo matching
- markov networks
- approximate inference
- fixed point
- pairwise
- loopy belief propagation
- computer vision
- energy function
- natural images
- image matching
- energy minimization
- free energy
- factor graphs
- d objects
- image features
- post processing
- probabilistic inference
- image retrieval
- three dimensional
- tree reweighted
- keypoints
- image segmentation