Hugging Face's logo

Shr3ezy
/
DenseFusion-ObjectPose

Model card Files
xet
Community
DenseFusion-ObjectPose
File size: 5,409 Bytes 
29858c0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
 
function plot_accuracy_keyframe

color = {'r', 'y', 'g', 'b', 'm'};
leng = {'iterative', 'PoseCNN+ICP', 'per-pixel', '3DCoordinate', ...
    '3D'};
aps = zeros(5, 1);
lengs = cell(5, 1);
close all;

% load results
object = load('results_keyframe.mat');
distances_sys = object.distances_sys;
distances_non = object.distances_non;
rotations = object.errors_rotation;
translations = object.errors_translation;
cls_ids = object.results_cls_id;

index_plot = [2, 3, 1, 5];

% read class names
fid = fopen('classes.txt', 'r');
C = textscan(fid, '%s');
classes = C{1};
classes{end+1} = 'All 21 objects';
fclose(fid);

hf = figure('units','normalized','outerposition',[0 0 1 1]);
font_size = 12;
max_distance = 0.1;

% for each class
for k = 1:numel(classes)
    index = find(cls_ids == k);
    if isempty(index)
        index = 1:size(distances_sys,1);
    end

    % distance symmetry
    subplot(2, 2, 1);
    for i = index_plot
        D = distances_sys(index, i);
        D(D > max_distance) = inf;
        d = sort(D);
        n = numel(d);
        c = numel(d(d < 0.02));
        accuracy = cumsum(ones(1, n)) / n;
%         fprintf('k = %d i = %d : length %d\n',k,i,length(d));
%         dd = find(d == d(end));
%         ddd = find(d ~= d(end));
%         fprintf('k = %d i = %d : length %d %d %d %d\n',k,i,length(d), length(dd), d(end), ddd(end));

        plot(d, accuracy, color{i}, 'LineWidth', 4);
        aps(i) = VOCap(d, accuracy);
        lengs{i} = sprintf('%s(AUC:%.2f)(<2cm:%.2f)', leng{i}, aps(i)*100, (c/n)*100);
        hold on;
    end
    hold off;
    %h = legend('network', 'refine tranlation only', 'icp', 'stereo translation only', 'stereo full', '3d coordinate');
    %set(h, 'FontSize', 16);
    h = legend(lengs(index_plot), 'Location', 'southeast');
    set(h, 'FontSize', font_size);
    h = xlabel('Average distance threshold in meter (symmetry)');
    set(h, 'FontSize', font_size);
    h = ylabel('accuracy');
    set(h, 'FontSize', font_size);
    h = title(classes{k}, 'Interpreter', 'none');
    set(h, 'FontSize', font_size);
    xt = get(gca, 'XTick');
    set(gca, 'FontSize', font_size)

    % distance non-symmetry
    subplot(2, 2, 2);
    for i = index_plot
        D = distances_non(index, i);
        D(D > max_distance) = inf;
        d = sort(D);
        n = numel(d);
        c = numel(d(d < 0.02));
        accuracy = cumsum(ones(1, n)) / n;
        plot(d, accuracy, color{i}, 'LineWidth', 4);
        aps(i) = VOCap(d, accuracy);
        lengs{i} = sprintf('%s(AUC:%.2f)(<2cm:%.2f)', leng{i}, aps(i)*100, (c/n)*100);        
        hold on;
    end
    hold off;
    %h = legend('network', 'refine tranlation only', 'icp', 'stereo translation only', 'stereo full', '3d coordinate');
    %set(h, 'FontSize', 16);
    h = legend(lengs(index_plot), 'Location', 'southeast');
    set(h, 'FontSize', font_size);
    h = xlabel('Average distance threshold in meter (non-symmetry)');
    set(h, 'FontSize', font_size);
    h = ylabel('accuracy');
    set(h, 'FontSize', font_size);
    h = title(classes{k}, 'Interpreter', 'none');
    set(h, 'FontSize', font_size);    
    xt = get(gca, 'XTick');
    set(gca, 'FontSize', font_size)
    
    % rotation
    subplot(2, 2, 3);
    for i = index_plot
        D = rotations(index, i);
        d = sort(D);
        n = numel(d);
        accuracy = cumsum(ones(1, n)) / n;
        plot(d, accuracy, color{i}, 'LineWidth', 4);
        hold on;
    end
    hold off;
    %h = legend('network', 'refine tranlation only', 'icp', 'stereo translation only', 'stereo full', '3d coordinate');
    %set(h, 'FontSize', 16);
    h = legend(leng(index_plot), 'Location', 'southeast');
    set(h, 'FontSize', font_size);
    h = xlabel('Rotation angle threshold');
    set(h, 'FontSize', font_size);
    h = ylabel('accuracy');
    set(h, 'FontSize', font_size);
    h = title(classes{k}, 'Interpreter', 'none');
    set(h, 'FontSize', font_size);
    xt = get(gca, 'XTick');
    set(gca, 'FontSize', font_size)

    % translation
    subplot(2, 2, 4);
    for i = index_plot
        D = translations(index, i);
        D(D > max_distance) = inf;
        d = sort(D);
        n = numel(d);
        accuracy = cumsum(ones(1, n)) / n;
        plot(d, accuracy, color{i}, 'LineWidth', 4);
        hold on;
    end
    hold off;
    h = legend(leng(index_plot), 'Location', 'southeast');
    set(h, 'FontSize', font_size);
    h = xlabel('Translation threshold in meter');
    set(h, 'FontSize', font_size);
    h = ylabel('accuracy');
    set(h, 'FontSize', font_size);
    h = title(classes{k}, 'Interpreter', 'none');
    set(h, 'FontSize', font_size);
    xt = get(gca, 'XTick');
    set(gca, 'FontSize', font_size)
    
    filename = sprintf('plots/%s.png', classes{k});
    hgexport(hf, filename, hgexport('factorystyle'), 'Format', 'png');
end

function ap = VOCap(rec, prec)

index = isfinite(rec);
rec = rec(index);
prec = prec(index)';



mrec=[0 ; rec ; 0.1];

% disp(prec)

% disp(end)

% disp(length(prec))

% if length(prec) == 0

%     prec(1) = 1;

% end

% disp(prec(end))



mpre=[0 ; prec ; prec(end)];

for i = 2:numel(mpre)

    mpre(i) = max(mpre(i), mpre(i-1));

end

i = find(mrec(2:end) ~= mrec(1:end-1)) + 1;

ap = sum((mrec(i) - mrec(i-1)) .* mpre(i)) * 10;