Download presentation

Presentation is loading. Please wait.

Published byNoah Sawyer Modified over 3 years ago

1
Mean-Field Theory and Its Applications In Computer Vision5 1

2
Global Co-occurrence Terms 2 Encourages global consistency and co- occurrence of objects Without cooc With co- occurrence

3
Global Co-occurrence Terms 3 Defined on subset of labels Associates a cost with each possible subset

4
Properties of cost function 4 Non-decreasing 0.2 3.0 5.0

5
Properties of cost function 5 We represent our cost as second order cost function defined on binary vector:

6
Complexity 6 Complexity: O(NL 2 ) Two relaxed (approximation) of this form Complexity: O(NL+L 2 )

7
Our model Represent 2 nd order cost by binary latent variables Unary cost per latent variable 7 label level variable node (0/1)

8
Our model Represent 2 nd order cost by binary latent variables Pairwise cost between latent variable 8

9
Global Co-occurrence Cost Two approximation to include into fully connected CRF 9

10
Global Co-occurrence Terms First model 10

11
Global Co-occurrence Terms Model 11

12
Global Co-occurrence Terms Constraints (lets take one set of connections) 12 If latent variable is on, atleast one of image variable take that label If latent variable is off, no image variable take that label

13
Global Co-occurrence Terms Pay a cost K for violating first constraint 13

14
Global Co-occurrence Terms Pay a cost K for violating second constrait 14

15
Global Co-occurrence Terms Cost for first model: 15

16
Global Co-occurrence Terms Second model Each latent node is connected to the variable node 16

17
Global Co-occurrence Terms Constraints (lets take one set of connections) 17 If latent variable is on, atleast one of image variable take that label If latent variable is off, no image variable take that label

18
Global Co-occurrence Terms Pay a cost K for violating the constraint 18

19
Global Co-occurrence Terms Cost for second model: 19

20
Global Co-occurrence Terms Expectation evaluation for variable Yl Case 1: Y_l takes label 0 20

21
Global Co-occurrence Terms Expectation evaluation for variable Yl Case 1: Y_l takes label 0 21

22
Global Co-occurrence Terms Expectation evaluation for variable Yl Case 1: Y_l takes label 0 22

23
Global Co-occurrence Terms Expectation evaluation for variable Yl Case 1: Y_l takes label 1 23

24
Global Co-occurrence Terms Expectation evaluation for variable Yl Case 1: Y_l takes label 1 24

25
Global Co-occurrence Terms Expectation evaluation for variable Yl 25

26
Global Co-occurrence Terms Latent variable updates: 26

27
Global Co-occurrence Terms Latent variable updates: 27

28
Global Co-occurrence Terms Pay a cost K if variable takes a label l and corresponding latent variable takes label 0 28

29
Complexity Expectation updates for latent variable Y_l 29

30
Complexity Expectation updates for latent variable Y_l 30 Overall complexity: Does not increase original complexity:

31
PascalVOC-10 dataset 31 Qualitative analysis: observe an improvement over other comparative methods

32
PascalVOC-10 dataset 32 AlgorithmTime (s)OverallAv. RecallAv. I/U AHCRF+Cooc3681.4338.0130.09 Dense CRF0.6771.4334.5328.40 Dense + Potts4.3579.8740.7130.18 Dense + Potts + Cooc 4.480.4443.0832.35 Observe an improvement of almost 2.3% improvement Almost 8-9 times faster than alpha-expansion based method

33
Mean-field Vs. Graph-cuts 33 Measure I/U score on PascalVOC-10 segmentation Increase standard deviation for mean-field Increase window size for graph-cuts method Both achieve almost similar accuracy

34
Window sizes 34 AlgorithmModelTime (s)Av. I/U Alpha-exp (n=10)Pairwise326.1728.59 Mean-fieldpairwise0.6728.64 Alpha-exp (n=3)Pairwise + Potts56.829.6 Mean-fieldPairwise + Potts4.3530.11 Alpha-exp (n=1)Pairwise + Potts + Cooc 103.9430.45 Mean-fieldPairwise + Potts + Cooc 4.432.17 Comparison on matched energy Impact of adding more complex costs and increasing window size

35
PascalVOC-10 dataset 35 AlgorithmbkgplaneCyclebirdBoat AHCRF+ Cooc 82.543.24.917.427.1 Dense + Potts + Cooc 82.944.615.818.926.3 AlgorithmbottleBuscarcatChair AHCRF+ Cooc 31.349.451.029.37.1 Dense + Potts + Cooc 31.748.955.233.37.9 Per class Quantitative results

36
PascalVOC-10 dataset 36 AlgorithmCowDtbdoghorseMbike AHCRF+ Cooc 26.78.317.024.027.1 Dense + Potts + Cooc 27.016.116.823.443.8 Algorith m psonPlantsheepsofatrainTVAv AHCRF+ Cooc 41.921.825.216.443.843.430.9 Dense + Potts + Cooc 38.421.130.915.544.036.832.35 Per class Quantitative results

37
Mean-field Vs. Graph-cuts 37 Measure I/U score on PascalVOC-10 segmentation Increase standard deviation for mean-field Increase window size for graph-cuts method Time complexity very high, making infeasible to work with large neighbourhood system

Similar presentations

OK

Budapest University of Technology and Economics, BME , 1872

Budapest University of Technology and Economics, BME , 1872

© 2017 SlidePlayer.com Inc.

All rights reserved.

Ads by Google

Ppt on noun for class 2 Ppt on channels of distribution of pepsi Ppt on introduction to object-oriented programming advantages Ppt on centering prayer Ppt on leadership qualities Ppt on object-oriented programming concepts c# Free ppt on customer service skills Ppt on earthquake for class 7 Ppt on brand management by keller How to present a seminar ppt on 4g