1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
|
%% Reset everything
clear all;
clc;
close all;
addpath('helpers');
%% Configure the benchmark
% central case -> only one camera
cam_number = 1;
% Getting 10 points, and testing all algorithms with the respective number of points
pt_number = 10;
% noise test, so no outliers
outlier_fraction = 0.0;
% repeat 5000 tests per noise level
iterations = 5000;
% The algorithms we want to test
algorithms = { 'fivept_stewenius'; 'fivept_nister'; 'fivept_kneip'; 'sevenpt'; 'eightpt'; 'eigensolver'; 'rel_nonlin_central' };
% Some parameter that tells us what the result means
returns = [ 1, 1, 0, 1, 1, 0, 2 ]; % 1means essential matrix(ces) needing decomposition, %0 means rotation matrix(ces), %2 means transformation matrix
% This defines the number of points used for every algorithm
indices = { [1, 2, 3, 4, 5]; [1, 2, 3, 4, 5]; [1, 2, 3, 4, 5]; [1, 2, 3, 4, 5, 6, 7]; [1, 2, 3, 4, 5, 6, 7, 8]; [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]; [1, 2, 3, 4, 5, 6, 7, 8, 9, 10] };
% The name of the algorithms in the final plots
names = { '5pt (Stewenius)'; '5pt (Nister)'; '5pt (Kneip)'; '7pt'; '8pt'; 'eigensolver (10pts)'; 'nonlin. opt. (10pts)' };
% The maximum noise to analyze
max_noise = 5.0;
% The step in between different noise levels
noise_step = 0.1;
%% Run the benchmark
%prepare the overall result arrays
number_noise_levels = max_noise / noise_step + 1;
num_algorithms = size(algorithms,1);
mean_rotation_errors = zeros(num_algorithms,number_noise_levels);
median_rotation_errors = zeros(num_algorithms,number_noise_levels);
noise_levels = zeros(1,number_noise_levels);
%Run the experiment
for n=1:number_noise_levels
noise = (n - 1) * noise_step;
noise_levels(1,n) = noise;
display(['Analyzing noise level: ' num2str(noise)])
rotation_errors = zeros(num_algorithms,iterations);
counter = 0;
validIterations = 0;
for i=1:iterations
% generate experiment
[v1,v2,t,R] = create2D2DExperiment(pt_number,cam_number,noise,outlier_fraction);
[t_perturbed,R_perturbed] = perturb(t,R,0.01);
T_perturbed = [R_perturbed,t_perturbed];
R_gt = R;
% run all algorithms
allValid = 1;
for a=1:num_algorithms
Out = opengv(algorithms{a},indices{a},v1,v2,T_perturbed);
if ~isempty(Out)
if returns(1,a) == 1
temp = transformEssentials(Out);
Out = temp;
end
if returns(1,a) == 2
temp = Out(:,1:3);
Out = temp;
end
rotation_errors(a,validIterations+1) = evaluateRotationError( R_gt, Out );
else
allValid = 0;
break;
end
end
if allValid == 1
validIterations = validIterations +1;
end
counter = counter + 1;
if counter == 100
counter = 0;
display(['Iteration ' num2str(i) ' of ' num2str(iterations) '(noise level ' num2str(noise) ')']);
end
end
%Now compute the mean and median value of the error for each algorithm
for a=1:num_algorithms
mean_rotation_errors(a,n) = mean(rotation_errors(a,1:validIterations));
median_rotation_errors(a,n) = median(rotation_errors(a,1:validIterations));
end
end
%% Plot the results
figure(1)
plot(noise_levels,mean_rotation_errors,'LineWidth',2)
legend(names,'Location','NorthWest')
xlabel('noise level [pix]')
ylabel('mean rot. error [rad]')
grid on
figure(2)
plot(noise_levels,median_rotation_errors,'LineWidth',2)
legend(names,'Location','NorthWest')
xlabel('noise level [pix]')
ylabel('median rot. error [rad]')
grid on
|
