-
Notifications
You must be signed in to change notification settings - Fork 142
/
fcn_test.m
executable file
·132 lines (110 loc) · 3.99 KB
/
fcn_test.m
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
124
125
126
127
128
129
130
131
132
close all; clear mex;
restoredefaultpath;
kCNNFolder = '.';
kMexFolder = fullfile(kCNNFolder, 'c++', 'build');
addpath(kMexFolder);
addpath('./matlab');
kModelPath = '/path/to/models';
load(fullfile(kModelPath, 'pascal-fcn32s-dag.mat'), 'params', 'meta');
PascalWeights = params;
MeanX = single(meta.normalization.averageImage);
ClassesNum = 21;
kImagePath = './data/pascal';
ImList = dir(fullfile(kImagePath, '*.jpg'));
confusion = zeros(ClassesNum);
for im = 1 : numel(ImList)
disp(im);
imname = ImList(im).name;
labname = strcat(imname(1:end-3), 'png');
if (~exist(fullfile(kImagePath, labname), 'file'))
continue;
end;
TestX = imread(fullfile(kImagePath, imname));
TestY = imread(fullfile(kImagePath, labname));
test_x = TestX;
kXSize = size(TestX);
kXSize(end+1:4) = 1;
test_x = single(test_x) - repmat(MeanX, [kXSize(1) kXSize(2) 1]);
TestY = single(TestY);
TestY = mod(TestY + 1, 256);
test_y = single(zeros(kXSize(1), kXSize(2), ClassesNum));
for c = 1 : ClassesNum
test_y(:, :, c) = (TestY == c);
end;
clear params;
params.epochs = 1;
params.alpha = 0.1;
params.beta = 0;
params.momentum = 0;
params.lossfun = 'logreg';
params.shuffle = 0;
params.seed = 1;
dropout = 0;
params.balance = 0;
params.verbose = 0;
params.gpu = 1;
layers = {
struct('type', 'input', 'mapsize', kXSize(1:2), 'channels', kXSize(3))
% group 1
struct('type', 'conv', 'filtersize', [3 3], 'channels', 64, 'padding', [100 100])
struct('type', 'conv', 'filtersize', [3 3], 'channels', 64, 'padding', [1 1])
struct('type', 'pool', 'scale', [2 2], 'stride', [2 2])
% group 2
struct('type', 'conv', 'filtersize', [3 3], 'channels', 128, 'padding', [1 1])
struct('type', 'conv', 'filtersize', [3 3], 'channels', 128, 'padding', [1 1])
struct('type', 'pool', 'scale', [2 2], 'stride', [2 2])
% group 3
struct('type', 'conv', 'filtersize', [3 3], 'channels', 256, 'padding', [1 1])
struct('type', 'conv', 'filtersize', [3 3], 'channels', 256, 'padding', [1 1])
struct('type', 'conv', 'filtersize', [3 3], 'channels', 256, 'padding', [1 1])
struct('type', 'pool', 'scale', [2 2], 'stride', [2 2])
% group 4
struct('type', 'conv', 'filtersize', [3 3], 'channels', 512, 'padding', [1 1])
struct('type', 'conv', 'filtersize', [3 3], 'channels', 512, 'padding', [1 1])
struct('type', 'conv', 'filtersize', [3 3], 'channels', 512, 'padding', [1 1])
struct('type', 'pool', 'scale', [2 2], 'stride', [2 2])
% group 5
struct('type', 'conv', 'filtersize', [3 3], 'channels', 512, 'padding', [1 1])
struct('type', 'conv', 'filtersize', [3 3], 'channels', 512, 'padding', [1 1])
struct('type', 'conv', 'filtersize', [3 3], 'channels', 512, 'padding', [1 1])
struct('type', 'pool', 'scale', [2 2], 'stride', [2 2])
% group 6
struct('type', 'conv', 'filtersize', [7 7], 'channels', 4096)
% group 7
struct('type', 'conv', 'filtersize', [1 1], 'channels', 4096)
% score2
struct('type', 'conv', 'filtersize', [1 1], 'channels', ClassesNum, ...
'function', 'none')
% upscore_s
struct('type', 'deconv', 'filtersize', [64 64], 'channels', ClassesNum, ...
'stride', [32 32], 'function', 'none')
% crop with softmax
struct('type', 'jitt', 'mapsize', kXSize(1:2), 'function', 'soft')
};
layers = import_weights(layers, PascalWeights, numel(layers));
weights = getweights(layers);
EpochNum = 1;
errors = zeros(EpochNum, 1);
for i = 1 : EpochNum
disp(['Epoch: ' num2str((i-1) * params.epochs + 1)]);
[err, bad, pred_y] = test(layers, weights, params, test_x, test_y);
[~, pred_ind] = max(pred_y, [], 3);
test_y = logical(test_y);
ok = TestY > 0;
confusion = confusion + accumarray([TestY(ok), pred_ind(ok)], 1, [21 21]);
subplot(2,2,1);
imshow(TestX);
subplot(2,2,2);
imshow(test_y(:, :, 1));
bg = pred_y(:, :, 1);
bg = (max(bg(:))-bg)/(max(bg(:))-min(bg(:)));
subplot(2,2,3);
imshow(bg);
[~, pred_ind] = max(pred_y, [], 3);
binary = (pred_ind(:, :, 1) == 2);
subplot(2,2,4);
imshow(binary);
end;
disp('Done!');
end;
[IU, meanIU, pixelAccuracy, meanAccuracy] = get_accuracies(confusion)