Download presentation
Presentation is loading. Please wait.
1
Belief Propagation Kai Ju Liu March 9, 2006
2
Statistical Problems Medicine Finance Internet Computer vision
3
Inference Problems Given data B, infer A: p(A|B) Computer vision –Given image, find objects –Given two images, resolve 3D object –Given multiple images, track object
4
Conditional Probability Given event B, what is probability of A? Independence: p(A|B)=p(A) AB
5
Bayes’ Rule
6
e.g. ColdWeekday Party Hangover Marginal Probability: 8-sum Joint Probability:
7
e.g. (cont.) ColdWeekday Party Hangover Marginal Probability: 8-sum Localize probabilities:
8
Approach Define variables and connections Calculate marginal probabilities efficiently Find most likely configuration
9
Pairwise Markov Random Field 123 4 5 Basic structure: vertices, edges
10
Pairwise Markov Random Field Basic structure: vertices, edges Vertex i has set of possible states X i and observed value y i Compatibility between states and observed values, Compatibility between neighboring vertices i and j,
11
Pairwise MRF: Probabilities Joint probability: Marginal probability: –Advantage: allows average over ambiguous states –Disadvantage: complexity exponential in number of vertices
12
Belief Propagation 123 4 5
13
Beliefs replace probabilities: Messages propagate information:
14
Belief Propagation Example 13 4 5
15
BP: Questions When can we calculate beliefs exactly? When do beliefs equal probabilities? When is belief propagation efficient? Answer: Singly-Connected Graphs (SCG’s) Graphs without loops Messages terminate at leaf vertices Beliefs equal probabilities Complexity in previous example reduced from 13S 5 to 24S 2
16
BP on Loopy Graphs Messages do not terminate Possible approximate solutions –Standard belief propagation –Generalized belief propagation BP-TwoGraphs: Goals Utilize advantages of SCG’s Be accurate and efficient on loopy graphs
17
BP-TwoGraphs: SCG’s Calculate beliefs on each set of SCG’s: – Select maximum beliefs from both sets – Consider loopy graph with n vertices Select two sets of SCG’s that approximate the graph –
18
BP-TwoGraphs: Vision SCG’s Rectangular grid of pixel vertices H i : horizontal graphs G i : vertical graphs
19
Image Segmentation add noise segment
20
Image Segmentation: Results
21
Real Image Segmentation
22
Real Image Segmentation: Training
23
Real Image Segmentation: Results
24
Stereo Vision
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.