open source pkg v1
This commit is contained in:
Vijay Yadev
@@ -0,0 +1,106 @@
|
||||
% collect the training data annotations (from the menpo challenge)
|
||||
clear
|
||||
train_data_loc = 'D:\Datasets\menpo/';
|
||||
|
||||
dataset_locs = {[train_data_loc, '/train/'];};
|
||||
|
||||
% Mapping all the points (profile and semi-frontal) to the same 68 point
|
||||
% annotation scheme
|
||||
left_to_frontal_map = [17,28; 18,29; 19,30; 20,31;
|
||||
21,34; 22,32; 23,39; 24,38; 25,37; 26,42; 27,41;
|
||||
28,52; 29,51; 30,50; 31,49; 32,60; 33,59; 34,58;
|
||||
35,63; 36,62; 37,61; 38,68; 39,67];
|
||||
|
||||
right_to_frontal_map = [17,28; 18,29; 19,30; 20,31;
|
||||
21,34; 22,36; 23,44; 24,45; 25,46; 26,47; 27,48;
|
||||
28,52; 29,53; 30,54; 31,55; 32,56; 33,57; 34,58;
|
||||
35,63; 36,64; 37,65; 38,66; 39,67];
|
||||
|
||||
all_pts = [];
|
||||
mirr_pts = [];
|
||||
|
||||
add_flip = true;
|
||||
|
||||
mirror_inds = [1,17;2,16;3,15;4,14;5,13;6,12;7,11;8,10;18,27;19,26;20,25;21,24;22,23;...
|
||||
32,36;33,35;37,46;38,45;39,44;40,43;41,48;42,47;49,55;50,54;51,53;60,56;59,57;...
|
||||
61,65;62,64;68,66];
|
||||
|
||||
for i=1:numel(dataset_locs)
|
||||
landmarkLabels = dir([dataset_locs{i} '\*.pts']);
|
||||
landmarkImgs = dir([dataset_locs{i} '\*.jpg']);
|
||||
|
||||
for p=1:numel(landmarkLabels)
|
||||
landmarks = importdata([dataset_locs{i}, landmarkLabels(p).name], ' ', 3);
|
||||
img = imread([dataset_locs{i}, landmarkImgs(p).name]);
|
||||
landmarks = landmarks.data;
|
||||
landmark_labels = -ones(68,2);
|
||||
|
||||
if(size(landmarks,1) == 39)
|
||||
% Determine if the points are clock-wise or counter clock-wise
|
||||
% Clock-wise points are facing left, counter-clock-wise right
|
||||
sum = 0;
|
||||
for k=1:11
|
||||
step = (landmarks(k+1,1) - landmarks(k,1)) * (landmarks(k+1,2) + landmarks(k,2));
|
||||
sum = sum + step;
|
||||
end
|
||||
|
||||
if(sum > 0)
|
||||
% First need to resample the face outline as there are 9
|
||||
% points in the near-frontal and 10 points in profile for
|
||||
% the outline of the face
|
||||
|
||||
outline = iterate_piece_wise(landmarks(1:10,:), 9);
|
||||
brow = iterate_piece_wise(landmarks(13:16,:), 5);
|
||||
landmark_labels(1:9,:) = outline;
|
||||
landmark_labels(18:22,:) = brow;
|
||||
landmark_labels(left_to_frontal_map(:,2),:) = landmarks(left_to_frontal_map(:,1),:);
|
||||
else
|
||||
outline = iterate_piece_wise(landmarks(10:-1:1,:), 9);
|
||||
brow = iterate_piece_wise(landmarks(16:-1:13,:), 5);
|
||||
|
||||
landmark_labels(9:17,:) = outline;
|
||||
landmark_labels(23:27,:) = brow;
|
||||
|
||||
landmark_labels(right_to_frontal_map(:,2),:) = landmarks(right_to_frontal_map(:,1),:);
|
||||
end
|
||||
else
|
||||
landmark_labels = landmarks;
|
||||
end
|
||||
|
||||
all_pts = cat(3, all_pts, landmark_labels);
|
||||
|
||||
if(add_flip)
|
||||
|
||||
mirror_lbls = landmark_labels;
|
||||
mirror_lbls(mirror_lbls ==-1) = nan;
|
||||
mirror_lbls(:,1) = -mirror_lbls(:,1);
|
||||
tmp1 = mirror_lbls(mirror_inds(:,1),:);
|
||||
tmp2 = mirror_lbls(mirror_inds(:,2),:);
|
||||
mirror_lbls(mirror_inds(:,2),:) = tmp1;
|
||||
mirror_lbls(mirror_inds(:,1),:) = tmp2;
|
||||
|
||||
mirror_lbls(isnan(mirror_lbls)) = -1;
|
||||
|
||||
mirr_pts = cat(3, mirr_pts, mirror_lbls);
|
||||
end
|
||||
|
||||
end
|
||||
end
|
||||
all_pts_old = all_pts;
|
||||
%%
|
||||
xs = squeeze(all_pts(:,1,:));
|
||||
ys = squeeze(all_pts(:,2,:));
|
||||
|
||||
all_pts = cat(2,xs,ys)';
|
||||
save('menpo_68_pts', 'all_pts');
|
||||
|
||||
%%
|
||||
all_pts = all_pts_old ;
|
||||
all_pts = cat(3, all_pts, mirr_pts);
|
||||
|
||||
xs = squeeze(all_pts(:,1,:));
|
||||
ys = squeeze(all_pts(:,2,:));
|
||||
|
||||
all_pts = cat(2,xs,ys)';
|
||||
|
||||
save('menpo_68_pts_flip', 'all_pts');
|
||||
@@ -0,0 +1,74 @@
|
||||
% collect the training data annotations (from the menpo challenge)
|
||||
clear
|
||||
train_data_loc = 'D:\Datasets\menpo/';
|
||||
|
||||
dataset_locs = {[train_data_loc, '/valid/'];};
|
||||
|
||||
|
||||
left_to_frontal_map = [17,28; 18,29; 19,30; 20,31;
|
||||
21,34; 22,32; 23,39; 24,38; 25,37; 26,42; 27,41;
|
||||
28,52; 29,51; 30,50; 31,49; 32,60; 33,59; 34,58;
|
||||
35,63; 36,62; 37,61; 38,68; 39,67];
|
||||
|
||||
right_to_frontal_map = [17,28; 18,29; 19,30; 20,31;
|
||||
21,34; 22,36; 23,44; 24,45; 25,46; 26,47; 27,48;
|
||||
28,52; 29,53; 30,54; 31,55; 32,56; 33,57; 34,58;
|
||||
35,63; 36,64; 37,65; 38,66; 39,67];
|
||||
|
||||
all_pts = [];
|
||||
|
||||
for i=1:numel(dataset_locs)
|
||||
landmarkLabels = dir([dataset_locs{i} '\*.pts']);
|
||||
landmarkImgs = dir([dataset_locs{i} '\*.jpg']);
|
||||
|
||||
for p=1:numel(landmarkLabels)
|
||||
|
||||
landmarks = importdata([dataset_locs{i}, landmarkLabels(p).name], ' ', 3);
|
||||
img = imread([dataset_locs{i}, landmarkImgs(p).name]);
|
||||
landmarks = landmarks.data;
|
||||
landmark_labels = -ones(68,2);
|
||||
|
||||
if(size(landmarks,1) == 39)
|
||||
% Determine if the points are clock-wise or counter clock-wise
|
||||
% Clock-wise points are facing left, counter-clock-wise right
|
||||
sum = 0;
|
||||
for k=1:11
|
||||
step = (landmarks(k+1,1) - landmarks(k,1)) * (landmarks(k+1,2) + landmarks(k,2));
|
||||
sum = sum + step;
|
||||
end
|
||||
|
||||
if(sum > 0)
|
||||
% First need to resample the face outline as there are 9
|
||||
% points in the near-frontal and 10 points in profile for
|
||||
% the outline of the face
|
||||
|
||||
outline = iterate_piece_wise(landmarks(1:10,:), 9);
|
||||
brow = iterate_piece_wise(landmarks(13:16,:), 5);
|
||||
landmark_labels(1:9,:) = outline;
|
||||
landmark_labels(18:22,:) = brow;
|
||||
landmark_labels(left_to_frontal_map(:,2),:) = landmarks(left_to_frontal_map(:,1),:);
|
||||
else
|
||||
outline = iterate_piece_wise(landmarks(10:-1:1,:), 9);
|
||||
brow = iterate_piece_wise(landmarks(16:-1:13,:), 5);
|
||||
|
||||
landmark_labels(9:17,:) = outline;
|
||||
landmark_labels(23:27,:) = brow;
|
||||
|
||||
landmark_labels(right_to_frontal_map(:,2),:) = landmarks(right_to_frontal_map(:,1),:);
|
||||
end
|
||||
else
|
||||
landmark_labels = landmarks;
|
||||
end
|
||||
|
||||
all_pts = cat(3, all_pts, landmark_labels);
|
||||
end
|
||||
end
|
||||
|
||||
%%
|
||||
|
||||
xs = squeeze(all_pts(:,1,:));
|
||||
ys = squeeze(all_pts(:,2,:));
|
||||
|
||||
all_pts = cat(2,xs,ys)';
|
||||
|
||||
save('menpo_68_pts_valid', 'all_pts');
|
||||
@@ -0,0 +1,101 @@
|
||||
clear;
|
||||
|
||||
addpath('../PDM_helpers/');
|
||||
|
||||
Reconstruct_Torresani;
|
||||
|
||||
%% Create the PDM using PCA, from the recovered 3D data
|
||||
clear
|
||||
load('Torr_menpo_wild.mat');
|
||||
|
||||
% need to still perform procrustes though
|
||||
|
||||
x = P3(1:end/3,:);
|
||||
y = P3(end/3+1:2*end/3,:);
|
||||
|
||||
% To make sure that PDM faces the right way (positive Z towards the screen)
|
||||
z = P3(2*end/3+1:end,:);
|
||||
|
||||
[ normX, normY, normZ, meanShape, Transform ] = ProcrustesAnalysis3D(x,y,z, true);
|
||||
observations = [normX normY normZ];
|
||||
|
||||
[princComp, score, eigenVals] = princomp(observations,'econ');
|
||||
% Keeping most variation
|
||||
totalSum = sum(eigenVals);
|
||||
count = numel(eigenVals);
|
||||
for i=1:numel(eigenVals)
|
||||
if ((sum(eigenVals(1:i)) / totalSum) >= 0.999)
|
||||
count = i;
|
||||
break;
|
||||
end
|
||||
end
|
||||
|
||||
V = princComp(:,1:count);
|
||||
E = eigenVals(1:count);
|
||||
M = meanShape(:);
|
||||
|
||||
% Now normalise it to have actual world scale
|
||||
lPupil = [(M(37) + M(40))/2; (M(37 + 68) + M(40 + 68))/2];
|
||||
rPupil = [(M(43) + M(46))/2; (M(43 + 68) + M(46 + 68))/2];
|
||||
|
||||
dist = norm(lPupil - rPupil,2);
|
||||
|
||||
% average human interocular distance is 65mm
|
||||
scaling = 65 / dist;
|
||||
|
||||
% normalise the mean values as well
|
||||
M(1:end/3) = M(1:end/3) - mean(M(1:end/3));
|
||||
M(end/3+1:2*end/3) = M(end/3+1:2*end/3) - mean(M(end/3+1:2*end/3));
|
||||
M(2*end/3+1:end) = M(2*end/3+1:end) - mean(M(2*end/3+1:end));
|
||||
|
||||
M = M * scaling;
|
||||
E = E * scaling .^ 2;
|
||||
|
||||
% orthonormalise V
|
||||
scalingV = sqrt(sum(V.^2));
|
||||
V = V ./ repmat(scalingV, numel(M),1);
|
||||
|
||||
E = E .* (scalingV') .^ 2;
|
||||
|
||||
%% Also align the model to a Multi-PIE one for accurate head orientation
|
||||
% also align the mean shape model (an aligned 68 point PDM from Multi-PIE dataset)
|
||||
M_wild = M;
|
||||
|
||||
load('pdm_68_multi_pie.mat', 'M');
|
||||
M_align = M;
|
||||
|
||||
n_points = numel(M)/3;
|
||||
|
||||
M_wild = reshape(M_wild, n_points, 3);
|
||||
|
||||
% work out the translation and rotation needed
|
||||
[ R, T ] = AlignShapesKabsch(M_wild, reshape(M_align, n_points,3));
|
||||
|
||||
% Transform the wild one to be in the same reference as the Multi-PIE one
|
||||
M_aligned = (R * M_wild')';
|
||||
|
||||
M_aligned(:,1) = M_aligned(:,1) + T(1);
|
||||
M_aligned(:,2) = M_aligned(:,2) + T(2);
|
||||
M_aligned(:,3) = M_aligned(:,3) + T(3);
|
||||
|
||||
M_aligned = M_aligned(:);
|
||||
V_aligned = V;
|
||||
|
||||
% need to align the principal components as well
|
||||
for i=1:size(V,2)
|
||||
|
||||
V_aligned_pie_curr = (R * reshape(V(:,i), n_points, 3)')';
|
||||
V_aligned(:,i) = V_aligned_pie_curr(:);
|
||||
|
||||
end
|
||||
|
||||
V = V_aligned;
|
||||
M = M_aligned;
|
||||
|
||||
save('pdm_68_aligned_menpo.mat', 'E', 'M', 'V');
|
||||
writePDM(V, E, M, 'pdm_68_aligned_menpo.txt');
|
||||
|
||||
% also save this to model location
|
||||
if(exist('../../models/pdm/', 'file'))
|
||||
save('../../models/pdm/pdm_68_aligned_menpo.mat', 'E', 'M', 'V');
|
||||
end
|
||||
@@ -0,0 +1,48 @@
|
||||
clear
|
||||
load('Noise_PDM/noise_0.2.mat');
|
||||
addpath('../PDM_helpers');
|
||||
|
||||
xs_orig = data_orig(:,1:end/2);
|
||||
ys_orig = data_orig(:,end/2+1:end);
|
||||
|
||||
xs_noise = data_noise(:,1:end/2);
|
||||
ys_noise = data_noise(:,end/2+1:end);
|
||||
|
||||
num_imgs = size(xs_noise, 1);
|
||||
|
||||
errs_orig = zeros(1,num_imgs);
|
||||
errs_noise = zeros(1,num_imgs);
|
||||
probs_orig = zeros(1,num_imgs);
|
||||
probs_noise = zeros(1,num_imgs);
|
||||
|
||||
pdmLoc = ['pdm_68_aligned_menpo_v5.mat'];
|
||||
|
||||
load(pdmLoc);
|
||||
|
||||
pdm = struct;
|
||||
pdm.M = double(M);
|
||||
pdm.E = double(E);
|
||||
pdm.V = double(V);
|
||||
|
||||
for i=1:num_imgs
|
||||
|
||||
labels_curr_orig = cat(2, xs_orig(i,:)', ys_orig(i,:)') * 100;
|
||||
labels_curr_orig(labels_curr_orig==-200) = 0;
|
||||
|
||||
[ a, R, T, ~, l_params, err, shapeOrtho] = fit_PDM_ortho_proj_to_2D(pdm.M, pdm.E, pdm.V, labels_curr_orig);
|
||||
errs_orig(i) = err/a;
|
||||
probs_orig(i) = sum(log(exp((-l_params.^2)./(pdm.E.^2))./(sqrt(2*pi.*pdm.E))));
|
||||
labels_curr_noise = cat(2, xs_noise(i,:)', ys_noise(i,:)') * 100;
|
||||
labels_curr_noise(labels_curr_orig==0) = 0;
|
||||
|
||||
[ a, R, T, ~, l_params, err, shapeOrtho] = fit_PDM_ortho_proj_to_2D(pdm.M, pdm.E, pdm.V, labels_curr_noise);
|
||||
errs_noise(i) = err/a;
|
||||
probs_noise(i) = sum(log(exp((-l_params.^2)./(pdm.E.^2))./(sqrt(2*pi.*pdm.E))));
|
||||
|
||||
end
|
||||
|
||||
% Current error is 2.1728 on the training data
|
||||
% After cleanup and smaller steps it is 1.5388
|
||||
% After training on it we have 1.1999
|
||||
% V2 - 1.1392
|
||||
% V3 - 1.1406
|
||||
@@ -0,0 +1,110 @@
|
||||
% This is the final model used for 3D PDM, it contains both Menpo and 300W
|
||||
% data
|
||||
clear
|
||||
|
||||
% Combining 300W and Menpo data for PDM
|
||||
load('menpo_68_pts_flip');
|
||||
num_pts = size(all_pts,1)/2;
|
||||
xs_m = all_pts(1:num_pts,:);
|
||||
ys_m = all_pts(num_pts+1:end,:);
|
||||
|
||||
load('wild_68_pts');
|
||||
num_pts = size(all_pts,1)/2;
|
||||
xs_w= all_pts(1:num_pts,:);
|
||||
ys_w = all_pts(num_pts+1:end,:);
|
||||
|
||||
all_pts = cat(1, xs_m, xs_w, ys_m, ys_w);
|
||||
|
||||
pdmLoc = ['../../models/pdm/pdm_68_aligned_wild.mat'];
|
||||
load(pdmLoc);
|
||||
% Before plugging into NRSFM, align the points in 2D
|
||||
num_pts = size(all_pts,1)/2;
|
||||
|
||||
num_lmks = numel(M) / 3;
|
||||
m = reshape(M, num_lmks, 3)';
|
||||
width_model = max(m(1,:)) - min(m(1,:));
|
||||
height_model = max(m(2,:)) - min(m(2,:));
|
||||
|
||||
for i=1:size(all_pts,1)/2
|
||||
|
||||
shape2D = cat(2, all_pts(i,:)', all_pts(i+num_pts,:)');
|
||||
|
||||
M_n = M;
|
||||
|
||||
if(sum(shape2D(:)==-1) > 0)
|
||||
|
||||
hidden = true;
|
||||
% which indices to remove
|
||||
inds_to_rem = shape2D(:,1) == -1 | shape2D(:,2) == -1;
|
||||
|
||||
shape2D = shape2D(~inds_to_rem,:);
|
||||
|
||||
inds_to_rem = repmat(inds_to_rem, 3, 1);
|
||||
|
||||
M_n = M(~inds_to_rem);
|
||||
|
||||
end
|
||||
|
||||
% To deal with really extreme cases of roll
|
||||
M2D = cat(2, M_n(1:end/3), M_n(end/3+1:2*end/3));
|
||||
[ A, t, error, alignedShape, s ] = AlignShapesWithScale(M2D, shape2D);
|
||||
R = A/s;
|
||||
|
||||
% Transform the shape
|
||||
shape2D(:,1) = shape2D(:,1) - t(1);
|
||||
shape2D(:,2) = shape2D(:,2) - t(2);
|
||||
|
||||
shape2D = (R' * shape2D')/s;
|
||||
shape2D = shape2D';
|
||||
|
||||
all_pts(i,all_pts(i,:)~=-1) = shape2D(:,1);
|
||||
all_pts(i+num_pts,all_pts(i+num_pts,:)~=-1) = shape2D(:,2);
|
||||
end
|
||||
|
||||
%%
|
||||
% to_rem = randperm(round(0.1*num_pts)); % Remove 10% to break some symmetry
|
||||
% all_pts([to_rem, to_rem+num_pts],:) = [];
|
||||
num_pts = size(all_pts,1)/2;
|
||||
|
||||
left_ids = all_pts(1:num_pts,10) == -1;
|
||||
right_ids = all_pts(1:num_pts,8) == -1;
|
||||
frontal = true(num_pts,1);
|
||||
frontal(left_ids | right_ids) = false;
|
||||
|
||||
%%
|
||||
xs = all_pts(1:num_pts,:);
|
||||
ys = all_pts(num_pts+1:end,:);
|
||||
|
||||
% Randperm the data, as a test
|
||||
|
||||
xs_f = xs(frontal,:);
|
||||
ys_f = ys(frontal,:);
|
||||
scatter(xs_f(:), -ys_f(:));
|
||||
|
||||
%% Perform NRSFM by Torresani
|
||||
addpath('../nrsfm-em');
|
||||
% (T is the number of frames, J is the number of points)
|
||||
|
||||
J = size(all_pts,2);
|
||||
T = size(all_pts,1)/2;
|
||||
|
||||
use_lds = 0; % not modeling a linear dynamic system here
|
||||
max_em_iter = 200;
|
||||
tol = 0.001;
|
||||
K = 30; % number of deformation shapes
|
||||
|
||||
MD = all_pts(1:end/2,:)==-1;
|
||||
|
||||
[P3, S_hat, V, RO, Tr, Z] = em_sfm(all_pts, MD, K, use_lds, tol, max_em_iter);
|
||||
|
||||
save('Torr_menpo_wild', 'P3', 'S_hat', 'V', 'RO', 'Tr', 'Z');
|
||||
|
||||
%%
|
||||
% xs = P3(1:num_pts,:);
|
||||
% ys = P3(num_pts+1:2*num_pts,:);
|
||||
% zs = P3(2*num_pts+1:end,:);
|
||||
% %
|
||||
% xs_f = xs(left_ids,:);
|
||||
% ys_f = ys(left_ids,:);
|
||||
% zs_f = zs(left_ids,:);
|
||||
% scatter3(xs_f(:), ys_f(:), zs_f(:));
|
||||
@@ -0,0 +1,97 @@
|
||||
clear
|
||||
|
||||
load('menpo_68_pts_flip');
|
||||
|
||||
pdmLoc = ['../../models/pdm/pdm_68_aligned_wild.mat'];
|
||||
load(pdmLoc);
|
||||
% Before plugging into NRSFM, align the points in 2D
|
||||
num_pts = size(all_pts,1)/2;
|
||||
|
||||
num_lmks = numel(M) / 3;
|
||||
m = reshape(M, num_lmks, 3)';
|
||||
width_model = max(m(1,:)) - min(m(1,:));
|
||||
height_model = max(m(2,:)) - min(m(2,:));
|
||||
|
||||
for i=1:size(all_pts,1)/2
|
||||
|
||||
shape2D = cat(2, all_pts(i,:)', all_pts(i+num_pts,:)');
|
||||
|
||||
M_n = M;
|
||||
|
||||
if(sum(shape2D(:)==-1) > 0)
|
||||
|
||||
hidden = true;
|
||||
% which indices to remove
|
||||
inds_to_rem = shape2D(:,1) == -1 | shape2D(:,2) == -1;
|
||||
|
||||
shape2D = shape2D(~inds_to_rem,:);
|
||||
|
||||
inds_to_rem = repmat(inds_to_rem, 3, 1);
|
||||
|
||||
M_n = M(~inds_to_rem);
|
||||
|
||||
end
|
||||
|
||||
% To deal with really extreme cases of roll
|
||||
M2D = cat(2, M_n(1:end/3), M_n(end/3+1:2*end/3));
|
||||
[ A, t, error, alignedShape, s ] = AlignShapesWithScale(M2D, shape2D);
|
||||
R = A/s;
|
||||
|
||||
% Transform the shape
|
||||
shape2D(:,1) = shape2D(:,1) - t(1);
|
||||
shape2D(:,2) = shape2D(:,2) - t(2);
|
||||
|
||||
shape2D = (R' * shape2D')/s;
|
||||
shape2D = shape2D';
|
||||
|
||||
all_pts(i,all_pts(i,:)~=-1) = shape2D(:,1);
|
||||
all_pts(i+num_pts,all_pts(i+num_pts,:)~=-1) = shape2D(:,2);
|
||||
end
|
||||
|
||||
%%
|
||||
% to_rem = randperm(round(0.1*num_pts)); % Remove 10% to break some symmetry
|
||||
% all_pts([to_rem, to_rem+num_pts],:) = [];
|
||||
num_pts = size(all_pts,1)/2;
|
||||
|
||||
left_ids = all_pts(1:num_pts,10) == -1;
|
||||
right_ids = all_pts(1:num_pts,8) == -1;
|
||||
frontal = true(num_pts,1);
|
||||
frontal(left_ids | right_ids) = false;
|
||||
|
||||
%%
|
||||
xs = all_pts(1:num_pts,:);
|
||||
ys = all_pts(num_pts+1:end,:);
|
||||
|
||||
% Randperm the data, as a test
|
||||
|
||||
% xs_f = xs(frontal,:);
|
||||
% ys_f = ys(frontal,:);
|
||||
% scatter(xs_f(:), -ys_f(:));
|
||||
|
||||
%% Perform NRSFM by Torresani
|
||||
addpath('../nrsfm-em');
|
||||
% (T is the number of frames, J is the number of points)
|
||||
|
||||
J = size(all_pts,2);
|
||||
T = size(all_pts,1)/2;
|
||||
|
||||
use_lds = 0; % not modeling a linear dynamic system here
|
||||
max_em_iter = 200;
|
||||
tol = 0.001;
|
||||
K = 30; % number of deformation shapes
|
||||
|
||||
MD = all_pts(1:end/2,:)==-1;
|
||||
|
||||
[P3, S_hat, V, RO, Tr, Z] = em_sfm(all_pts, MD, K, use_lds, tol, max_em_iter);
|
||||
|
||||
save('Torr_menpo', 'P3', 'S_hat', 'V', 'RO', 'Tr', 'Z');
|
||||
|
||||
%%
|
||||
% xs = P3(1:num_pts,:);
|
||||
% ys = P3(num_pts+1:2*num_pts,:);
|
||||
% zs = P3(2*num_pts+1:end,:);
|
||||
% %
|
||||
% xs_f = xs(left_ids,:);
|
||||
% ys_f = ys(left_ids,:);
|
||||
% zs_f = zs(left_ids,:);
|
||||
% scatter3(xs_f(:), ys_f(:), zs_f(:));
|
||||
163
pkg/OpenFace/model_training/pdm_generation/menpo_pdm/em_sfm.m
Normal file
163
pkg/OpenFace/model_training/pdm_generation/menpo_pdm/em_sfm.m
Normal file
@@ -0,0 +1,163 @@
|
||||
function [P3, S_bar, V, RO, Tr, Z, sigma_sq, phi, Q, mu0, sigma0] = em_sfm(P, MD, K, use_lds, tol, max_em_iter)
|
||||
|
||||
% Non-Rigid Structure From Motion with Gaussian/LDS Deformation Model
|
||||
% Copyright (c) by Lorenzo Torresani, Stanford University
|
||||
%
|
||||
% Based on the following paper:
|
||||
%
|
||||
% Lorenzo Torresani, Aaron Hertzmann and Christoph Bregler,
|
||||
% Learning Non-Rigid 3D Shape from 2D Motion, NIPS 16, 2003
|
||||
% http://cs.stanford.edu/~ltorresa/projects/learning-nr-shape/
|
||||
%
|
||||
% Please refer to this publication if you use this program for
|
||||
% research or for technical applications.
|
||||
%
|
||||
%
|
||||
% INPUT:
|
||||
%
|
||||
% P - (2*T) x J tracking matrix: P([t t+T],:) contains the 2D projections of the J points at time t
|
||||
% MD - T x J missing data binary matrix: MD(t, j)=1 if no valid data is available for point j at time t, 0 otherwise
|
||||
% K - number of deformation basis
|
||||
% use_lds - set to 1 to model deformations using a linear dynamical system; set to 0 otherwise
|
||||
% tol - termination tolerance (proportional change in likelihood)
|
||||
% max_em_iter - maximum number of EM iterations
|
||||
%
|
||||
%
|
||||
% OUTPUT:
|
||||
%
|
||||
% P3 - (3*T) x J 3D-motion matrix: ( P3([t t+T t+2*T],:) contains the 3D coordinates of the J points at time t )
|
||||
% S_bar - shape average: 3 x J matrix
|
||||
% V - deformation shapes: (3*K) x J matrix ( V((n-1)*3+[1:3],:) contains the n-th deformation basis )
|
||||
% RO - rotation: cell array ( RO{t} gives the rotation matrix at time t )
|
||||
% Tr - translation: T x 2 matrix
|
||||
% Z - deformation weights: T x K matrix
|
||||
% sigma_sq - variance of the noise in feature position
|
||||
% phi - LDS transition matrix
|
||||
% Q - LDS state noise matrix
|
||||
% mu0 - initial state mean
|
||||
% sigma0 - initial state variance
|
||||
|
||||
if mod(size(P,1), 1) ~= 0,
|
||||
fprintf('Error: size(P) must be (2*T)xJ\n');
|
||||
return;
|
||||
end
|
||||
if (size(P,1)/2 ~= size(MD,1)) | (size(P,2) ~= size(MD,2))
|
||||
fprintf('Error: Size incompatibility between P and MD\n');
|
||||
return;
|
||||
end
|
||||
if mod(K, 1) ~= 0,
|
||||
fprintf('Error: K must be an integer value\n');
|
||||
return;
|
||||
end
|
||||
|
||||
[T, J] = size(MD);
|
||||
r = 3*(K + 1); % motion rank
|
||||
|
||||
P_hat = P; % if any of the points are missing, P_hat will be updated during the M-step
|
||||
|
||||
% uses rank 3 factorization to get a first initialization for rotation and S_bar
|
||||
[R_init, Trvect, S_bar] = rigidfac(P_hat, MD);
|
||||
|
||||
Tr(:,1) = Trvect(1:T);
|
||||
Tr(:,2) = Trvect(T+1:2*T);
|
||||
|
||||
R = zeros(2*T, 3);
|
||||
% enforces rotation constraints
|
||||
for t = 1:T,
|
||||
Ru = R_init(t,:);
|
||||
Rv = R_init(T+t,:);
|
||||
Rz = cross(Ru,Rv); if det([Ru;Rv;Rz])<0, Rz = -Rz; end;
|
||||
RO_approx = apprRot([Ru;Rv;Rz]);
|
||||
RO{t} = RO_approx;
|
||||
R(t,:) = RO_approx(1,:);
|
||||
R(t+T,:) = RO_approx(2,:);
|
||||
end;
|
||||
|
||||
% given the initial estimates of rotation, translation and shape average, it initializes
|
||||
% deformation shapes and weights through LSQ minimization of the reprojection error
|
||||
[V, Z] = init_SB(P_hat, Tr, R, S_bar, K);
|
||||
|
||||
% initializes sigma_sq
|
||||
E_zz_init = cov(Z);
|
||||
E_zz_init = repmat(E_zz_init, T, 1);
|
||||
sigma_sq = mstep_update_noisevar(P_hat, S_bar, V, Z', E_zz_init, RO, Tr);
|
||||
|
||||
if use_lds,
|
||||
[phi, mu0, sigma0, Q] = init_lds(P_hat, S_bar, V, R, Tr, sigma_sq);
|
||||
else
|
||||
phi = [];
|
||||
mu0 = [];
|
||||
sigma0 = [];
|
||||
Q = [];
|
||||
end
|
||||
|
||||
loglik = 0;
|
||||
% annealing_const = 60;
|
||||
annealing_const = 100;
|
||||
|
||||
max_anneal_iter = round(max_em_iter/2);
|
||||
|
||||
for em_iter=1:max_em_iter,
|
||||
if use_lds,
|
||||
[E_z, E_zz, y, M, xt_n, Pt_n, Ptt1_n, xt_t1, Pt_t1] = estep_lds_compute_Z_distr(P_hat, S_bar, V, R, Tr, phi, mu0, sigma0, Q, sigma_sq);
|
||||
|
||||
[phi, Q, sigma_sq, mu0, sigma0] = mstep_lds_update(y, M, xt_n, Pt_n, Ptt1_n);
|
||||
else
|
||||
% computes the hidden variables distributions
|
||||
[E_z, E_zz] = estep_compute_Z_distr(P_hat, S_bar, V, R, Tr, sigma_sq); % (Eq 17-18)
|
||||
end
|
||||
Z = E_z';
|
||||
|
||||
% updates shape basis
|
||||
[S_bar, V] = mstep_update_shapebasis(P_hat, E_z, E_zz, R, Tr, S_bar, V); % (Eq 21)
|
||||
|
||||
% fills in missing points
|
||||
if sum(MD(:))>0,
|
||||
P_hat = mstep_update_missingdata(P_hat, MD, S_bar, V, E_z, RO, Tr); % (Eq 25)
|
||||
end
|
||||
|
||||
% updates rotation
|
||||
[RO, R] = mstep_update_rotation(P_hat, S_bar, V, E_z, E_zz, RO, Tr); % (Eq 24)
|
||||
|
||||
% updates translation
|
||||
Tr = mstep_update_transl(P_hat, S_bar, V, E_z, RO); % (Eq 23)
|
||||
|
||||
if ~use_lds,
|
||||
% updates noise variance
|
||||
sigma_sq = mstep_update_noisevar(P_hat, S_bar, V, E_z, E_zz, RO, Tr); % (Eq 22)
|
||||
if em_iter < max_anneal_iter,
|
||||
sigma_sq = sigma_sq * (1 + annealing_const*(1 - em_iter/max_anneal_iter));
|
||||
end
|
||||
|
||||
oldloglik = loglik;
|
||||
% computes log likelihood
|
||||
loglik = compute_log_lik(P_hat, S_bar, V, E_z, E_zz, RO, Tr, sigma_sq);
|
||||
|
||||
fprintf('LogLik(%d): %f\n', em_iter, loglik);
|
||||
|
||||
if (em_iter <= 2),
|
||||
loglikbase = loglik;
|
||||
elseif (loglik < oldloglik)
|
||||
fprintf('Violation');
|
||||
% keyboard;
|
||||
elseif 0 & ((loglik-loglikbase)<(1 + tol)*(oldloglik-loglikbase)),
|
||||
fprintf('\n');
|
||||
break;
|
||||
end
|
||||
else
|
||||
fprintf('Iteration %d/%d\n', em_iter, max_em_iter);
|
||||
end
|
||||
end
|
||||
|
||||
P3 = zeros(3*T, J);
|
||||
for t = 1:T,
|
||||
z_t = Z(t,:);
|
||||
Rf = [R(t,:); R(t+T,:)];
|
||||
S = S_bar;
|
||||
for kk = 1:K,
|
||||
S = S+z_t(kk)*V((kk-1)*3+[1:3],:);
|
||||
end;
|
||||
S = RO{t}*S;
|
||||
|
||||
P3([t t+T t+2*T], :) = S + [Tr(t, [1 2]) -mean(S(3,:))]'*ones(1,J);
|
||||
end
|
||||
@@ -0,0 +1,373 @@
|
||||
function [ a, R, T, T3D, params, error, shapeOrtho ] = fit_PDM_ortho_proj_to_2D( M, E, V, shape2D, f, cx, cy)
|
||||
%FITPDMTO2DSHAPE Summary of this function goes here
|
||||
% Detailed explanation goes here
|
||||
|
||||
params = zeros(size(E));
|
||||
|
||||
hidden = false;
|
||||
|
||||
% if some of the points are unavailable modify M, V, and shape2D (can
|
||||
% later infer the actual shape from this)
|
||||
if(sum(shape2D(:)==0) > 0)
|
||||
|
||||
hidden = true;
|
||||
% which indices to remove
|
||||
inds_to_rem = shape2D(:,1) == 0 | shape2D(:,2) == 0;
|
||||
|
||||
shape2D = shape2D(~inds_to_rem,:);
|
||||
|
||||
inds_to_rem = repmat(inds_to_rem, 3, 1);
|
||||
|
||||
M_old = M;
|
||||
V_old = V;
|
||||
|
||||
M = M(~inds_to_rem);
|
||||
V = V(~inds_to_rem,:);
|
||||
|
||||
end
|
||||
|
||||
num_points = numel(M) / 3;
|
||||
|
||||
m = reshape(M, num_points, 3)';
|
||||
width_model = max(m(1,:)) - min(m(1,:));
|
||||
height_model = max(m(2,:)) - min(m(2,:));
|
||||
|
||||
bounding_box = [min(shape2D(:,1)), min(shape2D(:,2)),...
|
||||
max(shape2D(:,1)), max(shape2D(:,2))];
|
||||
|
||||
a = (((bounding_box(3) - bounding_box(1)) / width_model) + ((bounding_box(4) - bounding_box(2))/ height_model)) / 2;
|
||||
|
||||
tx = (bounding_box(3) + bounding_box(1))/2;
|
||||
ty = (bounding_box(4) + bounding_box(2))/2;
|
||||
|
||||
% correct it so that the bounding box is just around the minimum
|
||||
% and maximum point in the initialised face
|
||||
tx = tx - a*(min(m(1,:)) + max(m(1,:)))/2;
|
||||
ty = ty - a*(min(m(2,:)) + max(m(2,:)))/2;
|
||||
|
||||
% To deal with really extreme cases of roll
|
||||
M3D = cat(2, M(1:end/3), M(end/3+1:2*end/3), zeros(numel(M(1:end/3)),1));
|
||||
shape3D = cat(2, shape2D, zeros(numel(M(1:end/3)),1));
|
||||
[ A, T, error, alignedShape, s ] = AlignShapesWithScale(M3D, shape3D);
|
||||
R = A/s;
|
||||
% R = eye(3);
|
||||
|
||||
T = [tx; ty];
|
||||
|
||||
currShape = getShapeOrtho(M, V, params, R, T, a);
|
||||
|
||||
currError = getRMSerror(currShape, shape2D);
|
||||
|
||||
reg_rigid = zeros(6,1);
|
||||
regFactor = 20;
|
||||
regularisations = [reg_rigid; regFactor ./ E]; % the above version, however, does not perform as well
|
||||
regularisations = diag(regularisations)*diag(regularisations);
|
||||
|
||||
red_in_a_row = 0;
|
||||
|
||||
for i=1:0
|
||||
|
||||
shape3D = M + V * params;
|
||||
shape3D = reshape(shape3D, numel(shape3D) / 3, 3);
|
||||
|
||||
% Now find the current residual error
|
||||
currShape = a * R(1:2,:)*shape3D' + repmat(T, 1, numel(M)/3);
|
||||
currShape = currShape';
|
||||
|
||||
error_res = shape2D - currShape;
|
||||
|
||||
eul = Rot2Euler(R);
|
||||
|
||||
p_global = [a; eul'; T];
|
||||
|
||||
% get the Jacobians
|
||||
J = CalcJacobian(M, V, params, p_global);
|
||||
|
||||
% RLMS style update
|
||||
p_delta = (J'*J + regularisations) \ (J'*error_res(:) - regularisations*[p_global;params]);
|
||||
|
||||
% Take smaller steps
|
||||
p_delta = 0.25 * p_delta;
|
||||
|
||||
[params, p_global] = CalcReferenceUpdate(p_delta, params, p_global);
|
||||
|
||||
% Make sure the params do not get out of hand
|
||||
params = ClampPDM(params, E);
|
||||
|
||||
a = p_global(1);
|
||||
R = Euler2Rot(p_global(2:4));
|
||||
T = p_global(5:6);
|
||||
|
||||
shape3D = M + V * params;
|
||||
shape3D = reshape(shape3D, numel(shape3D) / 3, 3);
|
||||
currShape = a * R(1:2,:)*shape3D' + repmat(T, 1, numel(M)/3);
|
||||
currShape = currShape';
|
||||
|
||||
error = getRMSerror(currShape, shape2D);
|
||||
|
||||
if(0.999 * currError < error)
|
||||
red_in_a_row = red_in_a_row + 1;
|
||||
if(red_in_a_row == 5)
|
||||
break;
|
||||
end
|
||||
end
|
||||
|
||||
currError = error;
|
||||
|
||||
end
|
||||
|
||||
if(hidden)
|
||||
shapeOrtho = getShapeOrtho(M_old, V_old, params, R, T, a);
|
||||
else
|
||||
shapeOrtho = currShape;
|
||||
end
|
||||
if(nargin == 7)
|
||||
|
||||
Zavg = f / a;
|
||||
Xavg = (T(1) - cx) / a;
|
||||
Yavg = (T(2) - cy) / a;
|
||||
|
||||
T3D = [Xavg;Yavg;Zavg];
|
||||
else
|
||||
T3D = [0;0;0];
|
||||
end
|
||||
|
||||
end
|
||||
|
||||
function [shape2D] = getShapeOrtho(M, V, p, R, T, a)
|
||||
|
||||
% M - mean shape vector
|
||||
% V - eigenvectors
|
||||
% p - parameters of non-rigid shape
|
||||
% R - rotation matrix
|
||||
% T - translation vector (tx, ty)
|
||||
shape3D = getShape3D(M, V, p);
|
||||
shape2D = a * R(1:2,:)*shape3D' + repmat(T, 1, numel(M)/3);
|
||||
shape2D = shape2D';
|
||||
end
|
||||
|
||||
function [shape2D] = getShapeOrthoFull(M, V, p, R, T, a)
|
||||
|
||||
% M - mean shape vector
|
||||
% V - eigenvectors
|
||||
% p - parameters of non-rigid shape
|
||||
% R - rotation matrix
|
||||
% T - translation vector (tx, ty)
|
||||
T = [T; 0];
|
||||
shape3D = getShape3D(M, V, p);
|
||||
shape2D = a * R*shape3D' + repmat(T, 1, numel(M)/3);
|
||||
shape2D = shape2D';
|
||||
end
|
||||
|
||||
function [shape3D] = getShape3D(M, V, params)
|
||||
|
||||
shape3D = M + V * params;
|
||||
shape3D = reshape(shape3D, numel(shape3D) / 3, 3);
|
||||
|
||||
end
|
||||
|
||||
function [error] = getRMSerror(shape2Dv1, shape2Dv2)
|
||||
|
||||
error = sqrt(mean(reshape(shape2Dv1 - shape2Dv2, numel(shape2Dv1), 1).^2));
|
||||
|
||||
end
|
||||
|
||||
% This calculates the combined rigid with non-rigid Jacobian
|
||||
function J = CalcJacobian(M, V, p, p_global)
|
||||
|
||||
n = size(M, 1)/3;
|
||||
|
||||
non_rigid_modes = size(V,2);
|
||||
|
||||
J = zeros(n*2, 6 + non_rigid_modes);
|
||||
|
||||
|
||||
% now the layour is
|
||||
% ---------- Rigid part -------------------|----Non rigid part--------|
|
||||
% dx_1/ds, dx_1/dr1, ... dx_1/dtx, dx_1/dty dx_1/dp_1 ... dx_1/dp_m
|
||||
% dx_2/ds, dx_2/dr1, ... dx_2/dtx, dx_2/dty dx_2/dp_1 ... dx_2/dp_m
|
||||
% ...
|
||||
% dx_n/ds, dx_n/dr1, ... dx_n/dtx, dx_n/dty dx_n/dp_1 ... dx_n/dp_m
|
||||
% dy_1/ds, dy_1/dr1, ... dy_1/dtx, dy_1/dty dy_1/dp_1 ... dy_1/dp_m
|
||||
% ...
|
||||
% dy_n/ds, dy_n/dr1, ... dy_n/dtx, dy_n/dty dy_n/dp_1 ... dy_n/dp_m
|
||||
|
||||
% getting the rigid part
|
||||
J(:,1:6) = CalcRigidJacobian(M, V, p, p_global);
|
||||
|
||||
% constructing the non-rigid part
|
||||
R = Euler2Rot(p_global(2:4));
|
||||
s = p_global(1);
|
||||
|
||||
% 'rotate' and 'scale' the principal components
|
||||
|
||||
% First reshape to 3D
|
||||
V_X = V(1:n,:);
|
||||
V_Y = V(n+1:2*n,:);
|
||||
V_Z = V(2*n+1:end,:);
|
||||
|
||||
J_x_non_rigid = s*(R(1,1)*V_X + R(1,2)*V_Y + R(1,3)*V_Z);
|
||||
J_y_non_rigid = s*(R(2,1)*V_X + R(2,2)*V_Y + R(2,3)*V_Z);
|
||||
|
||||
J(1:n, 7:end) = J_x_non_rigid;
|
||||
J(n+1:end, 7:end) = J_y_non_rigid;
|
||||
|
||||
end
|
||||
|
||||
function J = CalcRigidJacobian(M, V, p, p_global)
|
||||
|
||||
n = size(M, 1)/3;
|
||||
|
||||
% Get the current 3D shape (not affected by global transform, as this
|
||||
% is how the Jacobian was derived (for derivation please see
|
||||
% ../derivations/orthoJacobian
|
||||
shape3D = GetShape3D(M, V, p);
|
||||
|
||||
% Get the rotation matrix corresponding to current global orientation
|
||||
R = Euler2Rot(p_global(2:4));
|
||||
s = p_global(1);
|
||||
|
||||
% Rigid Jacobian is laid out as follows
|
||||
% dx_1/ds, dx_1/dr1, dx_1/dr2, dx_1/dr3, dx_1/dtx, dx_1/dty
|
||||
% dx_2/ds, dx_2/dr1, dx_2/dr2, dx_2/dr3, dx_2/dtx, dx_2/dty
|
||||
% ...
|
||||
% dx_n/ds, dx_n/dr1, dx_n/dr2, dx_n/dr3, dx_n/dtx, dx_n/dty
|
||||
% dy_1/ds, dy_1/dr1, dy_1/dr2, dy_1/dr3, dy_1/dtx, dy_1/dty
|
||||
% ...
|
||||
% dy_n/ds, dy_n/dr1, dy_n/dr2, dy_n/dr3, dy_n/dtx, dy_n/dty
|
||||
|
||||
J = zeros(n*2, 6);
|
||||
|
||||
% dx/ds = X * r11 + Y * r12 + Z * r13
|
||||
% dx/dr1 = s*(r13 * Y - r12 * Z)
|
||||
% dx/dr2 = -s*(r13 * X - r11 * Z)
|
||||
% dx/dr3 = s*(r12 * X - r11 * Y)
|
||||
% dx/dtx = 1
|
||||
% dx/dty = 0
|
||||
|
||||
% dy/ds = X * r21 + Y * r22 + Z * r23
|
||||
% dy/dr1 = s * (r23 * Y - r22 * Z)
|
||||
% dy/dr2 = -s * (r23 * X - r21 * Z)
|
||||
% dy/dr3 = s * (r22 * X - r21 * Y)
|
||||
% dy/dtx = 0
|
||||
% dy/dty = 1
|
||||
|
||||
% set the Jacobian for x's
|
||||
|
||||
% with respect to scaling factor
|
||||
J(1:n,1) = shape3D * R(1,:)';
|
||||
|
||||
% with respect to angular rotation around x, y, and z axes
|
||||
|
||||
% Change of x with respect to change in axis angle rotation
|
||||
dxdR = [ 0, R(1,3), -R(1,2);
|
||||
-R(1,3), 0, R(1,1);
|
||||
R(1,2), -R(1,1), 0];
|
||||
|
||||
J(1:n,2:4) = s*(dxdR * shape3D')';
|
||||
|
||||
% with respect to translation
|
||||
J(1:n,5) = 1;
|
||||
J(1:n,6) = 0;
|
||||
|
||||
% set the Jacobian for y's
|
||||
|
||||
% with respect to scaling factor
|
||||
J(n+1:end,1) = shape3D * R(2,:)';
|
||||
|
||||
% with respect to angular rotation around x, y, and z axes
|
||||
|
||||
% Change of y with respect to change in axis angle rotation
|
||||
dydR = [ 0, R(2,3), -R(2,2);
|
||||
-R(2,3), 0, R(2,1);
|
||||
R(2,2), -R(2,1), 0];
|
||||
|
||||
J(n+1:end,2:4) = s*(dydR * shape3D')';
|
||||
|
||||
% with respect to translation
|
||||
J(n+1:end,5) = 0;
|
||||
J(n+1:end,6) = 1;
|
||||
|
||||
end
|
||||
|
||||
% This updates the parameters based on the updates from the RLMS
|
||||
function [non_rigid, rigid] = CalcReferenceUpdate(params_delta, current_non_rigid, current_global)
|
||||
|
||||
|
||||
rigid = zeros(6, 1);
|
||||
% Same goes for scaling and translation parameters
|
||||
rigid(1) = current_global(1) + params_delta(1);
|
||||
rigid(5) = current_global(5) + params_delta(5);
|
||||
rigid(6) = current_global(6) + params_delta(6);
|
||||
|
||||
% for rotation however, we want to make sure that the rotation matrix
|
||||
% approximation we have
|
||||
% R' = [1, -wz, wy
|
||||
% wz, 1, -wx
|
||||
% -wy, wx, 1]
|
||||
% is a legal rotation matrix, and then we combine it with current
|
||||
% rotation (through matrix multiplication) to acquire the new rotation
|
||||
|
||||
R = Euler2Rot(current_global(2:4));
|
||||
|
||||
wx = params_delta(2);
|
||||
wy = params_delta(3);
|
||||
wz = params_delta(4);
|
||||
|
||||
R_delta = [1, -wz, wy;
|
||||
wz, 1, -wx;
|
||||
-wy, wx, 1];
|
||||
|
||||
% Make sure R_delta is orthonormal
|
||||
R_delta = double(OrthonormaliseRotation(R_delta));
|
||||
|
||||
% Combine rotations
|
||||
R_final = R * R_delta;
|
||||
|
||||
% Extract euler angle
|
||||
euler = real(Rot2Euler(R_final));
|
||||
|
||||
rigid(2:4) = euler;
|
||||
|
||||
if(length(params_delta) > 6)
|
||||
% non-rigid parameters can just be added together
|
||||
non_rigid = params_delta(7:end) + current_non_rigid;
|
||||
else
|
||||
non_rigid = current_non_rigid;
|
||||
end
|
||||
|
||||
end
|
||||
|
||||
function R_ortho = OrthonormaliseRotation(R)
|
||||
|
||||
% U * V' is basically what we want, as it's guaranteed to be
|
||||
% orthonormal
|
||||
[U, ~, V] = svd(R);
|
||||
|
||||
% We also want to make sure no reflection happened
|
||||
|
||||
% get the orthogonal matrix from the initial rotation matrix
|
||||
X = U*V';
|
||||
|
||||
% This makes sure that the handedness is preserved and no reflection happened
|
||||
% by making sure the determinant is 1 and not -1
|
||||
W = eye(3);
|
||||
W(3,3) = det(X);
|
||||
R_ortho = U*W*V';
|
||||
end
|
||||
|
||||
% This clamps the non-rigid parameters to stay within +- 3 standard
|
||||
% deviations
|
||||
function [non_rigid_params] = ClampPDM(non_rigid, E)
|
||||
|
||||
stds = sqrt(E);
|
||||
|
||||
non_rigid_params = non_rigid;
|
||||
|
||||
lower = non_rigid_params < -3 * stds;
|
||||
non_rigid_params(lower) = -3*stds(lower);
|
||||
|
||||
higher = non_rigid_params > 3 * stds;
|
||||
non_rigid_params(higher) = 3*stds(higher);
|
||||
|
||||
end
|
||||
@@ -0,0 +1,34 @@
|
||||
% A script for visualizing the orientations in the menpo data, allows to
|
||||
% see the distributions of the data
|
||||
clear
|
||||
load('../menpo_68_pts.mat');
|
||||
addpath('../../PDM_helpers');
|
||||
|
||||
xs = all_pts(1:end/2,:);
|
||||
ys = all_pts(end/2+1:end,:);
|
||||
num_imgs = size(xs, 1);
|
||||
|
||||
rots = zeros(3, num_imgs);
|
||||
errs = zeros(1,num_imgs);
|
||||
|
||||
pdmLoc = ['../pdm_68_aligned_menpo.mat'];
|
||||
|
||||
load(pdmLoc);
|
||||
|
||||
pdm = struct;
|
||||
pdm.M = double(M);
|
||||
pdm.E = double(E);
|
||||
pdm.V = double(V);
|
||||
errs_poss = [];
|
||||
for i=1:num_imgs
|
||||
|
||||
labels_curr = cat(2, xs(i,:)', ys(i,:)');
|
||||
labels_curr(labels_curr==-1) = 0;
|
||||
|
||||
[ a, R, T, ~, l_params, err, shapeOrtho] = fit_PDM_ortho_proj_to_2D(pdm.M, pdm.E, pdm.V, labels_curr);
|
||||
errs(i) = err/a;
|
||||
rots(:,i) = Rot2Euler(R);
|
||||
|
||||
end
|
||||
|
||||
hist(rots', 100);
|
||||
@@ -0,0 +1,121 @@
|
||||
function [images, detections, labels] = Collect_wild_imgs_train(root_test_data)
|
||||
|
||||
use_lfpw = true;
|
||||
use_helen = true;
|
||||
|
||||
use_68 = true;
|
||||
|
||||
images = [];
|
||||
labels = [];
|
||||
detections = [];
|
||||
|
||||
if(use_lfpw)
|
||||
[img, det, lbl] = Collect_LFPW(root_test_data, use_68);
|
||||
images = cat(1, images, img');
|
||||
detections = cat(1, detections, det);
|
||||
labels = cat(1, labels, lbl);
|
||||
end
|
||||
|
||||
if(use_helen)
|
||||
[img, det, lbl] = Collect_helen(root_test_data, use_68);
|
||||
images = cat(1, images, img');
|
||||
detections = cat(1, detections, det);
|
||||
labels = cat(1, labels, lbl);
|
||||
end
|
||||
|
||||
% convert to format expected by the Fitting method
|
||||
detections(:,3) = detections(:,1) + detections(:,3);
|
||||
detections(:,4) = detections(:,2) + detections(:,4);
|
||||
|
||||
|
||||
end
|
||||
|
||||
function [images, detections, labels] = Collect_LFPW(root_test_data, use_68)
|
||||
|
||||
dataset_loc = [root_test_data, '/lfpw/trainset/'];
|
||||
|
||||
landmarkLabels = dir([dataset_loc '\*.pts']);
|
||||
|
||||
num_imgs = size(landmarkLabels,1);
|
||||
|
||||
images = struct;
|
||||
if(use_68)
|
||||
labels = zeros(num_imgs, 68, 2);
|
||||
else
|
||||
labels = zeros(num_imgs, 66, 2);
|
||||
end
|
||||
|
||||
detections = zeros(num_imgs, 4);
|
||||
|
||||
load([root_test_data, '/Bounding Boxes/bounding_boxes_lfpw_trainset.mat']);
|
||||
|
||||
num_landmarks = 68;
|
||||
|
||||
for imgs = 1:num_imgs
|
||||
|
||||
[~,name,~] = fileparts(landmarkLabels(imgs).name);
|
||||
|
||||
landmarks = dlmread([dataset_loc, landmarkLabels(imgs).name], ' ', [3,0,num_landmarks+2,1]);
|
||||
|
||||
if(~use_68)
|
||||
inds_frontal = [1:60,62:64,66:68];
|
||||
landmarks = landmarks(inds_frontal,:);
|
||||
end
|
||||
|
||||
images(imgs).img = [dataset_loc, name '.png'];
|
||||
|
||||
labels(imgs,:,:) = landmarks;
|
||||
|
||||
detections(imgs,:) = bounding_boxes{imgs}.bb_detector;
|
||||
|
||||
end
|
||||
|
||||
detections(:,3) = detections(:,3) - detections(:,1);
|
||||
detections(:,4) = detections(:,4) - detections(:,2);
|
||||
|
||||
|
||||
end
|
||||
|
||||
function [images, detections, labels] = Collect_helen(root_test_data, use_68)
|
||||
|
||||
dataset_loc = [root_test_data, '/helen/trainset/'];
|
||||
|
||||
landmarkLabels = dir([dataset_loc '\*.pts']);
|
||||
|
||||
num_imgs = size(landmarkLabels,1);
|
||||
|
||||
images = struct;
|
||||
|
||||
if(use_68)
|
||||
labels = zeros(num_imgs, 68, 2);
|
||||
else
|
||||
labels = zeros(num_imgs, 66, 2);
|
||||
end
|
||||
|
||||
detections = zeros(num_imgs, 4);
|
||||
|
||||
load([root_test_data, '/Bounding Boxes/bounding_boxes_helen_trainset.mat']);
|
||||
num_landmarks = 68;
|
||||
for imgs = 1:num_imgs
|
||||
|
||||
[~,name,~] = fileparts(landmarkLabels(imgs).name);
|
||||
|
||||
landmarks = dlmread([dataset_loc, landmarkLabels(imgs).name], ' ', [3,0,num_landmarks+2,1]);
|
||||
|
||||
if(~use_68)
|
||||
inds_frontal = [1:60,62:64,66:68];
|
||||
landmarks = landmarks(inds_frontal,:);
|
||||
end
|
||||
|
||||
images(imgs).img = [dataset_loc, name '.jpg'];
|
||||
|
||||
labels(imgs,:,:) = landmarks;
|
||||
|
||||
detections(imgs,:) = bounding_boxes{imgs}.bb_detector;
|
||||
|
||||
end
|
||||
|
||||
detections(:,3) = detections(:,3) - detections(:,1);
|
||||
detections(:,4) = detections(:,4) - detections(:,2);
|
||||
|
||||
end
|
||||
@@ -0,0 +1,54 @@
|
||||
clear
|
||||
load('../menpo_68_pts_valid.mat');
|
||||
addpath('../../PDM_helpers');
|
||||
|
||||
xs = all_pts(1:end/2,:);
|
||||
ys = all_pts(end/2+1:end,:);
|
||||
num_imgs = size(xs, 1);
|
||||
|
||||
rots_menpo = zeros(3, num_imgs);
|
||||
errs_menpo = zeros(1,num_imgs);
|
||||
|
||||
pdmLoc = ['../pdm_68_aligned_menpo.mat'];
|
||||
|
||||
load(pdmLoc);
|
||||
|
||||
pdm = struct;
|
||||
pdm.M = double(M);
|
||||
pdm.E = double(E);
|
||||
pdm.V = double(V);
|
||||
errs_poss = [];
|
||||
for i=1:num_imgs
|
||||
|
||||
labels_curr = cat(2, xs(i,:)', ys(i,:)');
|
||||
labels_curr(labels_curr==-1) = 0;
|
||||
|
||||
[ a, R, T, ~, l_params, err, shapeOrtho] = fit_PDM_ortho_proj_to_2D(pdm.M, pdm.E, pdm.V, labels_curr);
|
||||
errs_menpo(i) = err/a;
|
||||
rots_menpo(:,i) = Rot2Euler(R);
|
||||
|
||||
if(errs_menpo(i) < 0 || errs_menpo(i) > 4)
|
||||
fprintf('i - %d, err - %.3f\n', i, errs_menpo(i));
|
||||
errs_poss = cat(1, errs_poss, i);
|
||||
end
|
||||
end
|
||||
|
||||
[~, ~, labels] = Collect_wild_imgs_train('D:/Dropbox/Dropbox/AAM/test data/');
|
||||
num_imgs = size(labels,1);
|
||||
rots_wild = zeros(3, num_imgs);
|
||||
errs_wild = zeros(1,num_imgs);
|
||||
|
||||
for i=1:num_imgs
|
||||
|
||||
labels_curr = squeeze(labels(i,:,:));
|
||||
labels_curr(labels_curr==-1) = 0;
|
||||
|
||||
[ a, R, T, ~, l_params, err, shapeOrtho] = fit_PDM_ortho_proj_to_2D(pdm.M, pdm.E, pdm.V, labels_curr);
|
||||
errs_wild(i) = err/a;
|
||||
rots_wild(:,i) = Rot2Euler(R);
|
||||
|
||||
end
|
||||
errs = cat(2, errs_wild, errs_menpo);
|
||||
fprintf('%.3f, %.3f, %.3f\n', mean(errs_menpo), mean(errs_wild), mean(errs));
|
||||
% 300W PDM error is 1.537, 1.017, 1.285
|
||||
% Menpo+300W PDM error is 1.130, 0.985, 1.060
|
||||
@@ -0,0 +1,157 @@
|
||||
% If a different type of PDM is to be trained, prepare training data here
|
||||
|
||||
clear
|
||||
load('menpo_68_pts_flip.mat');
|
||||
addpath('../PDM_helpers');
|
||||
|
||||
xs = all_pts(1:end/2,:);
|
||||
ys = all_pts(end/2+1:end,:);
|
||||
num_imgs = size(xs, 1);
|
||||
|
||||
pdmLoc = ['../../models/pdm/pdm_68_aligned_wild.mat'];
|
||||
% pdmLoc = ['pdm_68_aligned_menpo_v1.mat'];
|
||||
|
||||
load(pdmLoc);
|
||||
|
||||
pdm = struct;
|
||||
pdm.M = double(M);
|
||||
pdm.E = double(E);
|
||||
pdm.V = double(V);
|
||||
errs_poss = [];
|
||||
|
||||
all_shapes = -200 * ones(num_imgs, 68 * 2);
|
||||
min_x = 0;
|
||||
max_x = 0;
|
||||
min_y = 0;
|
||||
max_y = 0;
|
||||
|
||||
for i=1:num_imgs
|
||||
|
||||
shape2D = cat(2, all_pts(i,:)', all_pts(i+num_imgs,:)');
|
||||
|
||||
M_n = M;
|
||||
ind_rem = find(shape2D(:,1) == -1);
|
||||
to_keep = setdiff(1:68, ind_rem);
|
||||
|
||||
if(sum(shape2D(:)==-1) > 0)
|
||||
|
||||
hidden = true;
|
||||
% which indices to remove
|
||||
inds_to_rem = shape2D(:,1) == -1 | shape2D(:,2) == -1;
|
||||
|
||||
shape2D = shape2D(~inds_to_rem,:);
|
||||
|
||||
inds_to_rem = repmat(inds_to_rem, 3, 1);
|
||||
|
||||
M_n = M(~inds_to_rem);
|
||||
|
||||
end
|
||||
|
||||
% To deal with really extreme cases of roll
|
||||
M2D = cat(2, M_n(1:end/3), M_n(end/3+1:2*end/3));
|
||||
[ A, t, error, alignedShape, s ] = AlignShapesWithScale(M2D, shape2D);
|
||||
R = A/s;
|
||||
|
||||
% Transform the shape
|
||||
shape2D(:,1) = shape2D(:,1) - t(1);
|
||||
shape2D(:,2) = shape2D(:,2) - t(2);
|
||||
|
||||
shape2D = (R' * shape2D')/s;
|
||||
shape2D = shape2D';
|
||||
|
||||
% all_pts(i,all_pts(i,:)~=-1) = shape2D(:,1);
|
||||
% all_pts(i+num_pts,all_pts(i+num_pts,:)~=-1) = shape2D(:,2);
|
||||
%
|
||||
all_shapes(i,[to_keep,to_keep+68]) = shape2D(:);
|
||||
|
||||
if(min(shape2D(:,1)) < min_x)
|
||||
min_x = min(shape2D(:,1));
|
||||
end
|
||||
if(min(shape2D(:,2)) < min_y)
|
||||
min_y = min(shape2D(:,2));
|
||||
end
|
||||
if(max(shape2D(:,1)) > max_x)
|
||||
max_x = max(shape2D(:,1));
|
||||
end
|
||||
if(max(shape2D(:,2)) > max_y)
|
||||
max_y = max(shape2D(:,2));
|
||||
end
|
||||
end
|
||||
|
||||
% Scaling and shifting the model so that all points are from -1 to 1
|
||||
MD = all_shapes == -200;
|
||||
|
||||
xs = 2 * (all_shapes(:,1:68) - min_x)/(max_x-min_x) - 1;
|
||||
ys = 2 * (all_shapes(:,69:end) - min_y)/(max_y - min_y) - 1;
|
||||
|
||||
all_shapes = cat(2, xs, ys);
|
||||
all_shapes(MD) = -2;
|
||||
|
||||
xs = all_shapes(:,1:68);
|
||||
ys = all_shapes(:,69:end);
|
||||
|
||||
save('menpo_train.mat', 'all_shapes');
|
||||
|
||||
%%
|
||||
load('menpo_68_pts_valid.mat');
|
||||
|
||||
xs = all_pts(1:end/2,:);
|
||||
ys = all_pts(end/2+1:end,:);
|
||||
num_imgs = size(xs, 1);
|
||||
|
||||
all_shapes = -200 * ones(num_imgs, 68 * 2);
|
||||
|
||||
for i=1:num_imgs
|
||||
|
||||
shape2D = cat(2, all_pts(i,:)', all_pts(i+num_imgs,:)');
|
||||
|
||||
M_n = M;
|
||||
ind_rem = find(shape2D(:,1) == -1);
|
||||
to_keep = setdiff(1:68, ind_rem);
|
||||
|
||||
if(sum(shape2D(:)==-1) > 0)
|
||||
|
||||
hidden = true;
|
||||
% which indices to remove
|
||||
inds_to_rem = shape2D(:,1) == -1 | shape2D(:,2) == -1;
|
||||
|
||||
shape2D = shape2D(~inds_to_rem,:);
|
||||
|
||||
inds_to_rem = repmat(inds_to_rem, 3, 1);
|
||||
|
||||
M_n = M(~inds_to_rem);
|
||||
|
||||
end
|
||||
|
||||
% To deal with really extreme cases of roll
|
||||
M2D = cat(2, M_n(1:end/3), M_n(end/3+1:2*end/3));
|
||||
[ A, t, error, alignedShape, s ] = AlignShapesWithScale(M2D, shape2D);
|
||||
R = A/s;
|
||||
|
||||
% Transform the shape
|
||||
shape2D(:,1) = shape2D(:,1) - t(1);
|
||||
shape2D(:,2) = shape2D(:,2) - t(2);
|
||||
|
||||
shape2D = (R' * shape2D')/s;
|
||||
shape2D = shape2D';
|
||||
|
||||
% all_pts(i,all_pts(i,:)~=-1) = shape2D(:,1);
|
||||
% all_pts(i+num_pts,all_pts(i+num_pts,:)~=-1) = shape2D(:,2);
|
||||
%
|
||||
all_shapes(i,[to_keep,to_keep+68]) = shape2D(:);
|
||||
|
||||
end
|
||||
|
||||
% Scaling and shifting the model so that all points are from -1 to 1
|
||||
MD = all_shapes == -200;
|
||||
|
||||
xs = 2 * (all_shapes(:,1:68) - min_x)/(max_x-min_x) - 1;
|
||||
ys = 2 * (all_shapes(:,69:end) - min_y)/(max_y - min_y) - 1;
|
||||
|
||||
all_shapes = cat(2, xs, ys);
|
||||
all_shapes(MD) = -2;
|
||||
|
||||
xs = all_shapes(:,1:68);
|
||||
ys = all_shapes(:,69:end);
|
||||
|
||||
save('menpo_valid.mat', 'all_shapes');
|
||||
@@ -0,0 +1,32 @@
|
||||
function [ pts_new ] = iterate_piece_wise( pts_orig, new_num_points )
|
||||
%ITERATE_PIECE_WISE Summary of this function goes here
|
||||
% Detailed explanation goes here
|
||||
|
||||
% Reinterpolate the new points, but make sure they are still on the
|
||||
% same lines
|
||||
num_orig = size(pts_orig,1);
|
||||
|
||||
% Divide the number of original segments by number of new segments
|
||||
step_size = (num_orig - 1) / (new_num_points-1);
|
||||
|
||||
pts_new = zeros(new_num_points,2);
|
||||
|
||||
% Clamp the beginning and end, as they will be the same
|
||||
pts_new(1,:) = pts_orig(1,:);
|
||||
pts_new(end,:) = pts_orig(end,:);
|
||||
|
||||
for i=1:new_num_points-2
|
||||
|
||||
low_point = floor(1 + i * step_size);
|
||||
high_point = ceil(1 + i * step_size);
|
||||
|
||||
coeff_1 = floor(1 + i * step_size) - i * step_size;
|
||||
coeff_2 = 1 - coeff_1;
|
||||
|
||||
new_pt = coeff_1 * pts_orig(low_point,:) + coeff_2 * pts_orig(high_point,:);
|
||||
pts_new(i+1,:) = new_pt;
|
||||
|
||||
end
|
||||
|
||||
end
|
||||
|
||||
Binary file not shown.
Binary file not shown.
Binary file not shown.
@@ -0,0 +1,77 @@
|
||||
% collect the training data annotations (from the menpo challenge)
|
||||
clear
|
||||
train_data_loc = 'C:\Users\tbaltrus\Documents\menpo_data_orig/';
|
||||
|
||||
dataset_locs = {[train_data_loc, '/train/'];};
|
||||
|
||||
|
||||
left_to_frontal_map = [17,28; 18,29; 19,30; 20,31;
|
||||
21,34; 22,32; 23,39; 24,38; 25,37; 26,42; 27,41;
|
||||
28,52; 29,51; 30,50; 31,49; 32,60; 33,59; 34,58;
|
||||
35,63; 36,62; 37,61; 38,68; 39,67];
|
||||
|
||||
right_to_frontal_map = [17,28; 18,29; 19,30; 20,31;
|
||||
21,34; 22,36; 23,44; 24,45; 25,46; 26,47; 27,48;
|
||||
28,52; 29,53; 30,54; 31,55; 32,56; 33,57; 34,58;
|
||||
35,63; 36,64; 37,65; 38,66; 39,67];
|
||||
|
||||
all_pts_left = [];
|
||||
all_pts_right = [];
|
||||
all_pts_orig_left = [];
|
||||
all_pts_orig_right = [];
|
||||
|
||||
for i=1:numel(dataset_locs)
|
||||
landmarkLabels = dir([dataset_locs{i} '\*.pts']);
|
||||
landmarkImgs = dir([dataset_locs{i} '\*.jpg']);
|
||||
|
||||
for p=1:numel(landmarkLabels)
|
||||
landmarks = importdata([dataset_locs{i}, landmarkLabels(p).name], ' ', 3);
|
||||
img = imread([dataset_locs{i}, landmarkImgs(p).name]);
|
||||
landmarks = landmarks.data;
|
||||
landmark_labels = -ones(68,2);
|
||||
|
||||
if(size(landmarks,1) == 39)
|
||||
% Determine if the points are clock-wise or counter clock-wise
|
||||
% Clock-wise points are facing left, counter-clock-wise right
|
||||
sum = 0;
|
||||
for k=1:11
|
||||
step = (landmarks(k+1,1) - landmarks(k,1)) * (landmarks(k+1,2) + landmarks(k,2));
|
||||
sum = sum + step;
|
||||
end
|
||||
|
||||
if(sum > 0)
|
||||
% First need to resample the face outline as there are 9
|
||||
% points in the near-frontal and 10 points in profile for
|
||||
% the outline of the face
|
||||
|
||||
outline = iterate_piece_wise(landmarks(1:10,:), 9);
|
||||
brow = iterate_piece_wise(landmarks(13:16,:), 5);
|
||||
landmark_labels(1:9,:) = outline;
|
||||
landmark_labels(18:22,:) = brow;
|
||||
landmark_labels(left_to_frontal_map(:,2),:) = landmarks(left_to_frontal_map(:,1),:);
|
||||
|
||||
all_pts_orig_right = cat(3, all_pts_orig_right, landmarks);
|
||||
all_pts_right = cat(3, all_pts_right, landmark_labels);
|
||||
else
|
||||
outline = iterate_piece_wise(landmarks(10:-1:1,:), 9);
|
||||
brow = iterate_piece_wise(landmarks(16:-1:13,:), 5);
|
||||
|
||||
landmark_labels(9:17,:) = outline;
|
||||
landmark_labels(23:27,:) = brow;
|
||||
|
||||
landmark_labels(right_to_frontal_map(:,2),:) = landmarks(right_to_frontal_map(:,1),:);
|
||||
|
||||
all_pts_orig_left = cat(3, all_pts_orig_left, landmarks);
|
||||
all_pts_left = cat(3, all_pts_left, landmark_labels);
|
||||
end
|
||||
end
|
||||
|
||||
if(mod(p,50) == 0)
|
||||
fprintf('%d/%d\n', p, numel(landmarkLabels))
|
||||
end
|
||||
end
|
||||
end
|
||||
|
||||
%%
|
||||
|
||||
save('menpo_68_pts_train_profile', 'all_pts_orig_right', 'all_pts_right', 'all_pts_orig_left', 'all_pts_left');
|
||||
@@ -0,0 +1,77 @@
|
||||
% collect the training data annotations (from the menpo challenge)
|
||||
clear
|
||||
train_data_loc = 'C:\Users\tbaltrus\Documents\menpo_data_orig/';
|
||||
|
||||
dataset_locs = {[train_data_loc, '/valid/'];};
|
||||
|
||||
|
||||
left_to_frontal_map = [17,28; 18,29; 19,30; 20,31;
|
||||
21,34; 22,32; 23,39; 24,38; 25,37; 26,42; 27,41;
|
||||
28,52; 29,51; 30,50; 31,49; 32,60; 33,59; 34,58;
|
||||
35,63; 36,62; 37,61; 38,68; 39,67];
|
||||
|
||||
right_to_frontal_map = [17,28; 18,29; 19,30; 20,31;
|
||||
21,34; 22,36; 23,44; 24,45; 25,46; 26,47; 27,48;
|
||||
28,52; 29,53; 30,54; 31,55; 32,56; 33,57; 34,58;
|
||||
35,63; 36,64; 37,65; 38,66; 39,67];
|
||||
|
||||
all_pts_left = [];
|
||||
all_pts_right = [];
|
||||
all_pts_orig_left = [];
|
||||
all_pts_orig_right = [];
|
||||
|
||||
for i=1:numel(dataset_locs)
|
||||
landmarkLabels = dir([dataset_locs{i} '\*.pts']);
|
||||
landmarkImgs = dir([dataset_locs{i} '\*.jpg']);
|
||||
|
||||
for p=1:numel(landmarkLabels)
|
||||
landmarks = importdata([dataset_locs{i}, landmarkLabels(p).name], ' ', 3);
|
||||
img = imread([dataset_locs{i}, landmarkImgs(p).name]);
|
||||
landmarks = landmarks.data;
|
||||
landmark_labels = -ones(68,2);
|
||||
|
||||
if(size(landmarks,1) == 39)
|
||||
% Determine if the points are clock-wise or counter clock-wise
|
||||
% Clock-wise points are facing left, counter-clock-wise right
|
||||
sum = 0;
|
||||
for k=1:11
|
||||
step = (landmarks(k+1,1) - landmarks(k,1)) * (landmarks(k+1,2) + landmarks(k,2));
|
||||
sum = sum + step;
|
||||
end
|
||||
|
||||
if(sum > 0)
|
||||
% First need to resample the face outline as there are 9
|
||||
% points in the near-frontal and 10 points in profile for
|
||||
% the outline of the face
|
||||
|
||||
outline = iterate_piece_wise(landmarks(1:10,:), 9);
|
||||
brow = iterate_piece_wise(landmarks(13:16,:), 5);
|
||||
landmark_labels(1:9,:) = outline;
|
||||
landmark_labels(18:22,:) = brow;
|
||||
landmark_labels(left_to_frontal_map(:,2),:) = landmarks(left_to_frontal_map(:,1),:);
|
||||
|
||||
all_pts_orig_right = cat(3, all_pts_orig_right, landmarks);
|
||||
all_pts_right = cat(3, all_pts_right, landmark_labels);
|
||||
else
|
||||
outline = iterate_piece_wise(landmarks(10:-1:1,:), 9);
|
||||
brow = iterate_piece_wise(landmarks(16:-1:13,:), 5);
|
||||
|
||||
landmark_labels(9:17,:) = outline;
|
||||
landmark_labels(23:27,:) = brow;
|
||||
|
||||
landmark_labels(right_to_frontal_map(:,2),:) = landmarks(right_to_frontal_map(:,1),:);
|
||||
|
||||
all_pts_orig_left = cat(3, all_pts_orig_left, landmarks);
|
||||
all_pts_left = cat(3, all_pts_left, landmark_labels);
|
||||
end
|
||||
end
|
||||
|
||||
if(mod(p,50) == 0)
|
||||
fprintf('%d/%d\n', p, numel(landmarkLabels))
|
||||
end
|
||||
end
|
||||
end
|
||||
|
||||
%%
|
||||
|
||||
save('menpo_68_pts_valid_profile', 'all_pts_orig_right', 'all_pts_right', 'all_pts_orig_left', 'all_pts_left');
|
||||
@@ -0,0 +1,34 @@
|
||||
function [ error_per_image, frontal_ids ] = compute_error_menpo_prof( ground_truth_all, detected_points_all )
|
||||
%compute_error
|
||||
% compute the average point-to-point Euclidean error normalized by the
|
||||
% inter-ocular distance (measured as the Euclidean distance between the
|
||||
% outer corners of the eyes)
|
||||
%
|
||||
% Inputs:
|
||||
% grounth_truth_all, size: num_of_points x 2 x num_of_images
|
||||
% detected_points_all, size: num_of_points x 2 x num_of_images
|
||||
% Output:
|
||||
% error_per_image, size: num_of_images x 1
|
||||
|
||||
|
||||
num_of_images = numel(ground_truth_all);
|
||||
error_per_image = zeros(num_of_images,1);
|
||||
frontal_ids = true(num_of_images,1);
|
||||
|
||||
for i =1:num_of_images
|
||||
detected_points = detected_points_all{i};
|
||||
ground_truth_points = ground_truth_all{i};
|
||||
num_of_points = size(ground_truth_points,1);
|
||||
|
||||
normalization = (max(ground_truth_points(:,1))-min(ground_truth_points(:,1)))+...
|
||||
(max(ground_truth_points(:,2))-min(ground_truth_points(:,2)));
|
||||
normalization = normalization / 2;
|
||||
|
||||
sum=0;
|
||||
for j=1:num_of_points
|
||||
sum = sum+norm(detected_points(j,:)-ground_truth_points(j,:));
|
||||
end
|
||||
error_per_image(i) = sum/(num_of_points*normalization);
|
||||
end
|
||||
|
||||
end
|
||||
Binary file not shown.
@@ -0,0 +1,75 @@
|
||||
clear
|
||||
load('menpo_68_pts_train_profile');
|
||||
|
||||
vis_pts_left = all_pts_left(:,1,1) ~= -1;
|
||||
xs = all_pts_left(vis_pts_left,:,:);
|
||||
xs = squeeze(cat(1, xs(:,1,:), xs(:,2,:)));
|
||||
|
||||
ys = squeeze(cat(1, all_pts_orig_left(:,1,:), all_pts_orig_left(:,2,:)));
|
||||
|
||||
a_left = ys /xs;
|
||||
a_left(abs(a_left)<0.001) = 0;
|
||||
transformed_left = a_left * xs;
|
||||
xs_pred_left = transformed_left(1:end/2,:);
|
||||
ys_pred_left = transformed_left(end/2+1:end,:);
|
||||
|
||||
vis_pts_right = all_pts_right(:,1,1) ~= -1;
|
||||
xs = all_pts_right(vis_pts_right,:,:);
|
||||
xs = squeeze(cat(1, xs(:,1,:), xs(:,2,:)));
|
||||
|
||||
ys = squeeze(cat(1, all_pts_orig_right(:,1,:), all_pts_orig_right(:,2,:)));
|
||||
|
||||
a_right = ys /xs;
|
||||
a_right(abs(a_right)<0.001) = 0;
|
||||
transformed_right = a_right * xs;
|
||||
xs_pred_right = transformed_right(1:end/2,:);
|
||||
ys_pred_right = transformed_right(end/2+1:end,:);
|
||||
% pred_left = cat(2,transformed_left
|
||||
|
||||
%%
|
||||
num = 110;
|
||||
plot(all_pts_orig_right(:,1,num), -all_pts_orig_right(:,2,num), '.r');
|
||||
hold on;
|
||||
plot(xs_pred_right(:,num), -ys_pred_right(:,num), '.b');
|
||||
hold off;
|
||||
|
||||
%% Evaluate on the validation data (see how much error is added)
|
||||
|
||||
load('menpo_68_pts_valid_profile');
|
||||
|
||||
xs = all_pts_left(vis_pts_left,:,:);
|
||||
xs = squeeze(cat(1, xs(:,1,:), xs(:,2,:)));
|
||||
|
||||
transformed_left = a_left * xs;
|
||||
xs_pred_left = transformed_left(1:end/2,:);
|
||||
ys_pred_left = transformed_left(end/2+1:end,:);
|
||||
|
||||
shapes = cell(size(ys_pred_left,1),1);
|
||||
labels = cell(size(ys_pred_left,1),1);
|
||||
|
||||
for i=1:numel(shapes)
|
||||
shapes{i} = cat(2, xs_pred_left(:,i), ys_pred_left(:,i));
|
||||
labels{i} = all_pts_orig_left(:,:,i);
|
||||
end
|
||||
|
||||
errors_left = compute_error_menpo_prof(labels, shapes);
|
||||
|
||||
|
||||
xs = all_pts_right(vis_pts_right,:,:);
|
||||
xs = squeeze(cat(1, xs(:,1,:), xs(:,2,:)));
|
||||
|
||||
transformed_right = a_right * xs;
|
||||
xs_pred_right = transformed_right(1:end/2,:);
|
||||
ys_pred_right = transformed_right(end/2+1:end,:);
|
||||
|
||||
shapes = cell(size(ys_pred_right,1),1);
|
||||
labels = cell(size(ys_pred_right,1),1);
|
||||
|
||||
for i=1:numel(shapes)
|
||||
shapes{i} = cat(2, xs_pred_right(:,i), ys_pred_right(:,i));
|
||||
labels{i} = all_pts_orig_right(:,:,i);
|
||||
end
|
||||
|
||||
errors_right = compute_error_menpo_prof(labels, shapes);
|
||||
save('conversion', 'a_left', 'a_right', 'vis_pts_left', 'vis_pts_right');
|
||||
% Try with scaling
|
||||
Binary file not shown.
Binary file not shown.
@@ -0,0 +1 @@
|
||||
This includes scripts necessary to learn the mapping from the detected points using our PDM to the points expected in menpo (the different face outline and chin points in profile).
|
||||
Binary file not shown.
@@ -0,0 +1,421 @@
|
||||
# The mean values of the components (in mm)
|
||||
204
|
||||
1
|
||||
6
|
||||
-74.409969
|
||||
-74.137773
|
||||
-72.056496
|
||||
-67.915820
|
||||
-60.492143
|
||||
-48.932581
|
||||
-34.280137
|
||||
-17.760628
|
||||
0.514393
|
||||
18.358848
|
||||
34.248759
|
||||
48.283175
|
||||
59.483187
|
||||
66.851931
|
||||
71.085235
|
||||
73.234417
|
||||
73.635334
|
||||
-57.908717
|
||||
-49.068078
|
||||
-37.656400
|
||||
-25.729901
|
||||
-14.795319
|
||||
14.837927
|
||||
25.725682
|
||||
37.468905
|
||||
48.751250
|
||||
57.344747
|
||||
0.388587
|
||||
0.295350
|
||||
0.206926
|
||||
0.135388
|
||||
-15.498362
|
||||
-8.047687
|
||||
0.076176
|
||||
8.251717
|
||||
15.588112
|
||||
-45.515042
|
||||
-37.079098
|
||||
-27.956753
|
||||
-19.734098
|
||||
-28.329816
|
||||
-37.307204
|
||||
19.581417
|
||||
27.839109
|
||||
36.880288
|
||||
45.145696
|
||||
37.224189
|
||||
28.382643
|
||||
-27.977066
|
||||
-17.476087
|
||||
-6.663955
|
||||
0.244789
|
||||
7.291085
|
||||
17.947928
|
||||
28.125385
|
||||
18.687933
|
||||
8.539212
|
||||
0.242857
|
||||
-7.847825
|
||||
-18.237373
|
||||
-24.528857
|
||||
-7.104831
|
||||
0.273636
|
||||
7.739429
|
||||
24.736241
|
||||
7.968382
|
||||
0.229350
|
||||
-7.397604
|
||||
-30.248138
|
||||
-11.542661
|
||||
7.567939
|
||||
26.259376
|
||||
42.866270
|
||||
56.586917
|
||||
67.359002
|
||||
75.143546
|
||||
77.529037
|
||||
75.071547
|
||||
67.393621
|
||||
56.803934
|
||||
43.251437
|
||||
26.710196
|
||||
8.051821
|
||||
-10.968992
|
||||
-29.588592
|
||||
-52.172718
|
||||
-59.222312
|
||||
-61.949207
|
||||
-61.037213
|
||||
-57.532329
|
||||
-57.980133
|
||||
-61.715278
|
||||
-62.824686
|
||||
-60.232398
|
||||
-53.273185
|
||||
-40.542060
|
||||
-30.162960
|
||||
-20.232369
|
||||
-9.953997
|
||||
3.145708
|
||||
4.856711
|
||||
5.687939
|
||||
4.792678
|
||||
2.997034
|
||||
-37.288250
|
||||
-42.119954
|
||||
-42.362645
|
||||
-36.532986
|
||||
-34.532435
|
||||
-34.563887
|
||||
-36.758354
|
||||
-42.734934
|
||||
-42.649721
|
||||
-37.900147
|
||||
-35.115277
|
||||
-34.873489
|
||||
27.684553
|
||||
22.191257
|
||||
18.679379
|
||||
20.310742
|
||||
18.619478
|
||||
22.094260
|
||||
27.477987
|
||||
35.483072
|
||||
39.911465
|
||||
40.897777
|
||||
40.075424
|
||||
35.645402
|
||||
27.883771
|
||||
25.627237
|
||||
26.352545
|
||||
25.545734
|
||||
27.760907
|
||||
31.094250
|
||||
32.006269
|
||||
31.195088
|
||||
55.141955
|
||||
51.629462
|
||||
48.507238
|
||||
44.913190
|
||||
38.574081
|
||||
29.347878
|
||||
17.493884
|
||||
5.304010
|
||||
-1.371218
|
||||
5.342776
|
||||
17.851699
|
||||
30.323049
|
||||
40.570487
|
||||
47.890182
|
||||
52.297654
|
||||
55.916658
|
||||
60.074025
|
||||
13.228242
|
||||
6.263352
|
||||
-0.547326
|
||||
-6.314237
|
||||
-10.909358
|
||||
-13.043499
|
||||
-9.974582
|
||||
-5.801725
|
||||
-0.545529
|
||||
5.625345
|
||||
-13.691691
|
||||
-22.845392
|
||||
-32.311712
|
||||
-41.584645
|
||||
-21.080547
|
||||
-26.294217
|
||||
-29.003103
|
||||
-26.743598
|
||||
-22.061192
|
||||
9.230938
|
||||
3.879971
|
||||
1.219552
|
||||
1.017579
|
||||
0.985420
|
||||
3.658622
|
||||
-0.023947
|
||||
-0.773850
|
||||
0.704347
|
||||
5.283591
|
||||
0.421741
|
||||
-1.190907
|
||||
-7.844583
|
||||
-17.491041
|
||||
-24.300421
|
||||
-25.186808
|
||||
-24.409816
|
||||
-17.587587
|
||||
-7.690400
|
||||
-15.385836
|
||||
-21.456027
|
||||
-22.469566
|
||||
-21.062494
|
||||
-15.340380
|
||||
-9.728696
|
||||
-21.188579
|
||||
-22.512963
|
||||
-21.275175
|
||||
-9.282340
|
||||
-20.462720
|
||||
-21.609047
|
||||
-20.300170
|
||||
# The principal components (eigenvectors) of identity or combined identity and expression model
|
||||
204
|
||||
30
|
||||
6
|
||||
-0.190469 -0.071184 -0.054685 0.088181 -0.005885 0.117801 -0.017233 -0.027212 -0.081376 -0.065794 0.104253 -0.043337 0.060539 0.027374 -0.030921 0.019905 -0.032263 0.073285 -0.027532 0.067513 -0.106417 -0.052799 0.090934 0.071815 -0.103673 -0.018281 -0.142612 -0.136544 0.104777 -0.059287
|
||||
-0.181987 -0.077301 -0.040768 0.081699 -0.010115 0.074060 -0.030833 -0.002966 -0.045981 -0.047412 0.056452 -0.056207 0.046458 0.008748 -0.018044 0.054030 -0.035939 0.107036 0.006886 0.061526 -0.048319 -0.066964 0.074145 0.045579 -0.068658 0.034547 -0.074267 -0.093346 -0.011087 -0.057005
|
||||
-0.173515 -0.088597 -0.030820 0.084278 -0.018017 0.017465 -0.045721 0.022537 -0.010900 -0.021144 0.014483 -0.051229 0.043695 0.007049 0.008784 0.079550 -0.021801 0.121824 0.044955 0.054364 0.003371 -0.068850 0.062924 0.019579 -0.006031 0.059153 0.004337 -0.040960 -0.127230 -0.069391
|
||||
-0.161922 -0.097905 -0.029552 0.085923 -0.024253 -0.051497 -0.041131 0.032695 0.013614 0.006737 -0.022233 -0.015979 0.044671 0.016456 0.042545 0.100738 0.010838 0.104136 0.059488 0.030909 0.041065 -0.005391 0.037132 0.000199 0.070025 0.027990 0.075888 0.013117 -0.161032 -0.045139
|
||||
-0.127025 -0.102943 -0.027545 0.074271 -0.037877 -0.128029 -0.020264 0.013901 0.046950 0.027499 -0.019909 0.036689 0.040723 0.031274 0.062878 0.095211 0.070627 0.052483 0.070186 0.002020 0.041430 0.113021 0.010653 -0.018427 0.107179 -0.036024 0.106632 0.053605 -0.091824 0.008829
|
||||
-0.073496 -0.095531 -0.016800 0.050067 -0.046629 -0.178622 -0.002550 -0.009981 0.073447 0.025813 0.031626 0.083468 0.033074 0.041344 0.067269 0.045120 0.119514 -0.030770 0.072274 -0.032900 0.021413 0.187191 -0.009658 -0.017608 0.058195 -0.076714 0.086848 0.093781 0.020418 0.041344
|
||||
-0.014336 -0.060707 -0.001135 0.024155 -0.031345 -0.180344 -0.005564 -0.018983 0.082040 -0.001669 0.104994 0.108331 0.040063 0.040731 0.054548 -0.031716 0.100364 -0.115422 0.062475 -0.077651 0.018518 0.152803 -0.023172 0.006727 -0.034741 -0.063271 0.027194 0.112111 0.091212 0.065332
|
||||
0.033288 -0.029650 0.010784 0.001490 -0.004227 -0.122205 -0.025230 -0.030788 0.060539 -0.045171 0.193095 0.125399 0.079600 0.061105 0.034746 -0.107021 -0.005095 -0.155007 0.019670 -0.119697 0.047390 0.049773 -0.078525 0.043392 -0.131234 -0.023157 -0.050902 0.107143 0.052837 0.100170
|
||||
0.042466 -0.023240 0.012901 -0.014628 0.013362 -0.027104 -0.033300 -0.036843 0.007515 -0.087840 0.228327 0.134957 0.110228 0.073733 0.007839 -0.116583 -0.133476 -0.130037 -0.026603 -0.136182 0.080216 -0.004143 -0.151400 0.000197 -0.162840 0.037012 -0.063016 0.066786 -0.057942 0.096117
|
||||
0.042275 -0.027210 0.015916 -0.021138 0.022148 0.080905 -0.031487 -0.040800 -0.036973 -0.122569 0.157395 0.099065 0.077464 0.014712 -0.019067 -0.059201 -0.177906 -0.067693 -0.025927 -0.123735 0.112479 -0.022961 -0.115332 -0.069161 -0.082514 0.084603 -0.009326 -0.012504 -0.166641 -0.008810
|
||||
0.034992 0.019866 0.016672 -0.026031 0.035146 0.156866 -0.035760 -0.037457 -0.044565 -0.148905 0.070856 0.043275 0.008939 -0.080557 -0.053465 -0.018797 -0.138466 -0.037546 0.015549 -0.104396 0.125055 -0.075197 0.001187 -0.063625 0.014932 0.119950 0.015538 -0.067638 -0.195703 -0.111059
|
||||
0.022083 0.076666 0.020893 -0.036598 0.041623 0.170524 -0.028483 -0.049185 -0.024061 -0.142174 0.012057 0.000080 -0.053553 -0.151003 -0.080378 -0.028844 -0.061412 -0.068752 0.050127 -0.077572 0.111086 -0.098424 0.097831 -0.028401 0.047053 0.120124 -0.009350 -0.076550 -0.101073 -0.108623
|
||||
0.000917 0.122188 0.018726 -0.053554 0.031875 0.131996 0.005304 -0.051658 -0.008332 -0.086419 -0.025484 -0.019507 -0.086669 -0.153646 -0.082325 -0.047700 0.026855 -0.100987 0.066018 -0.042915 0.060076 -0.037594 0.127929 -0.011704 0.032725 0.075172 -0.046255 -0.036993 0.042759 -0.027430
|
||||
-0.015849 0.147622 0.013988 -0.065844 0.023894 0.060816 0.039563 -0.042881 -0.004454 -0.003180 -0.031554 -0.009564 -0.079995 -0.090976 -0.044620 -0.046412 0.085008 -0.104415 0.056154 -0.006056 -0.006052 0.072131 0.082830 -0.023428 0.003381 0.005341 -0.066595 -0.015071 0.173112 0.078035
|
||||
-0.024307 0.153924 0.015329 -0.068699 0.026456 -0.007556 0.048898 -0.038035 -0.007825 0.073291 -0.016401 0.010907 -0.047848 -0.002683 0.023197 -0.014666 0.080274 -0.065237 0.024079 0.024507 -0.060286 0.128755 0.010244 -0.064376 -0.013535 -0.037682 -0.039532 -0.001064 0.183772 0.090049
|
||||
-0.036264 0.160721 0.024597 -0.068751 0.026582 -0.065025 0.032341 -0.033475 0.017568 0.128566 0.002109 0.018173 -0.014675 0.052460 0.083642 0.014227 0.048078 -0.019013 -0.012220 0.058774 -0.073209 0.109006 -0.064971 -0.100035 -0.020187 -0.043025 0.008835 0.007521 0.093237 0.015679
|
||||
-0.046696 0.169704 0.036840 -0.076586 0.027841 -0.112892 0.008739 -0.031926 0.053716 0.165163 0.020429 0.017699 0.003555 0.079056 0.125320 0.034425 0.005298 0.024924 -0.042072 0.097605 -0.059220 0.081114 -0.133639 -0.125511 -0.037272 -0.028366 0.062529 0.011166 -0.004243 -0.068946
|
||||
0.042291 0.076030 -0.093282 0.067198 0.007215 0.103402 -0.006997 -0.104492 -0.121120 0.016173 -0.008727 0.004814 0.089177 -0.186746 0.011826 0.085202 -0.062554 -0.004803 -0.155804 -0.071192 0.136436 0.102667 -0.034988 0.081817 -0.079375 -0.139207 -0.036352 -0.004241 0.065533 -0.070118
|
||||
0.052443 0.056537 -0.072345 0.060115 -0.011066 0.080535 0.003019 -0.127387 -0.129043 0.086946 -0.006090 0.118992 0.033195 -0.174404 0.046975 0.039567 -0.034670 -0.049415 -0.144558 -0.041391 0.114681 0.080133 -0.040871 0.019499 0.009455 -0.054976 0.022552 0.063668 -0.004017 -0.084620
|
||||
0.047463 0.014756 -0.043425 0.051131 -0.020178 0.076732 -0.013818 -0.126892 -0.079018 0.146515 0.035490 0.158779 -0.107349 -0.105306 0.048792 -0.027827 -0.045394 -0.053009 -0.102846 -0.022955 0.069459 0.057348 -0.009400 -0.001550 0.084086 0.050101 0.049623 0.056422 -0.023817 -0.046569
|
||||
0.048007 -0.024661 -0.015357 0.038002 -0.013363 0.076925 -0.040654 -0.114865 -0.010085 0.183850 0.073473 0.136677 -0.237996 -0.021463 0.034689 -0.088979 -0.089578 -0.022381 -0.064666 -0.022433 0.026354 0.058752 0.041014 0.004505 0.142270 0.148051 0.035265 0.038667 -0.008270 -0.007768
|
||||
0.051793 -0.056949 0.013723 0.022806 -0.005622 0.072354 -0.068065 -0.092880 0.061062 0.204887 0.090190 0.089521 -0.311144 0.053182 0.022133 -0.132114 -0.136610 0.024535 -0.040255 -0.035124 -0.015640 0.080175 0.084529 0.006064 0.173201 0.219697 0.005246 0.039355 0.007063 0.018692
|
||||
0.048551 0.012281 0.000290 -0.028771 0.025829 -0.088107 0.006261 -0.091056 -0.034144 -0.191198 -0.047651 -0.077389 0.359607 -0.025798 -0.056870 0.035983 0.055101 -0.032975 -0.068922 0.057166 -0.046956 0.104589 0.095031 -0.085005 0.182852 0.257621 0.107048 0.058628 -0.015172 0.073623
|
||||
0.007636 0.001957 0.019717 -0.034560 0.014475 -0.086702 -0.015406 -0.080933 0.068802 -0.195883 0.013528 -0.084509 0.246520 -0.043931 -0.056795 0.048090 0.051544 -0.019811 -0.082600 0.076202 0.011676 0.066159 0.082000 -0.049743 0.149956 0.134417 0.078092 0.054942 -0.012085 0.015554
|
||||
-0.032833 -0.011149 0.036685 -0.038468 0.000077 -0.082118 -0.023279 -0.075403 0.157067 -0.188836 0.065133 -0.077001 0.088676 -0.042061 -0.061185 0.056591 0.038765 -0.019920 -0.081673 0.088792 0.072993 -0.017075 0.067129 -0.018861 0.116261 -0.018749 0.036306 0.017326 -0.008303 -0.012383
|
||||
-0.071594 -0.029101 0.056928 -0.040265 -0.029169 -0.085809 -0.009701 -0.072193 0.214929 -0.141362 0.092370 -0.037606 -0.063282 -0.008065 -0.069296 0.039686 0.033955 -0.027633 -0.047596 0.103709 0.111083 -0.136264 0.027231 -0.010179 0.075506 -0.157117 0.011433 -0.023525 -0.026937 -0.031782
|
||||
-0.102902 -0.026645 0.071626 -0.042690 -0.060351 -0.106379 0.034893 -0.067981 0.195447 -0.075360 0.062952 0.051391 -0.105644 0.030290 -0.052445 0.017990 0.040063 -0.047170 0.013324 0.085582 0.104803 -0.234707 -0.056129 -0.043783 0.044242 -0.217184 0.020767 -0.040826 -0.064307 -0.083017
|
||||
0.073954 -0.033481 0.011348 -0.013852 0.025753 -0.011342 -0.038623 -0.041194 -0.014761 0.040073 0.005668 -0.013876 0.029282 0.054587 0.016941 -0.027225 -0.003344 0.045073 0.053799 -0.010496 -0.011153 0.027733 0.027007 -0.050946 -0.012546 0.028573 -0.020853 0.021921 -0.036245 -0.005055
|
||||
0.066281 -0.038531 0.010994 -0.010512 0.020687 -0.001944 -0.022969 -0.018157 -0.021210 0.038141 -0.006614 -0.020176 0.026154 0.054925 0.012944 -0.036839 0.007008 0.042922 0.088378 -0.009251 0.051186 0.043168 0.060672 -0.080082 -0.018936 -0.014647 -0.033627 0.036186 -0.029742 -0.008131
|
||||
0.060145 -0.045291 0.011506 -0.007381 0.015724 0.007401 -0.008034 0.002664 -0.026686 0.036472 -0.014297 -0.025851 0.024334 0.057238 0.010284 -0.047580 0.016618 0.038845 0.116682 -0.003316 0.111662 0.059627 0.093922 -0.109972 -0.023774 -0.055271 -0.047776 0.051498 -0.021311 -0.009858
|
||||
0.056953 -0.054151 0.013023 -0.004828 0.011346 0.016032 0.006524 0.023821 -0.032325 0.034980 -0.020104 -0.031507 0.021464 0.059350 0.007546 -0.057002 0.026837 0.033675 0.139803 0.005629 0.167677 0.074225 0.127121 -0.138162 -0.026789 -0.093805 -0.061564 0.068483 -0.012682 -0.008529
|
||||
0.039049 0.004331 0.009235 0.037546 0.049086 -0.014903 -0.025946 0.048417 0.002055 0.023134 -0.020108 -0.066426 0.008865 0.016069 0.004197 -0.017642 0.020358 -0.009648 0.157395 0.134970 0.175886 -0.174092 0.076699 -0.071553 -0.044112 0.040272 0.037466 0.216854 0.252999 -0.172764
|
||||
0.040038 -0.007091 0.007822 0.021372 0.034784 -0.005405 -0.006969 0.041693 -0.009682 0.028397 -0.020225 -0.048332 0.001538 0.029769 -0.001635 -0.036032 0.019107 -0.005341 0.133339 0.091366 0.152670 -0.085737 0.089831 -0.058282 -0.034860 -0.021070 -0.005042 0.156095 0.149845 -0.083745
|
||||
0.035957 -0.031162 0.005084 -0.001774 0.007203 0.012038 0.013807 0.035198 -0.033169 0.027079 -0.040665 -0.032007 0.014259 0.037937 -0.001077 -0.043557 0.026802 0.014658 0.081716 0.027843 0.109242 0.034435 0.086752 -0.083719 -0.017238 -0.091026 -0.047042 0.035921 0.012064 0.021800
|
||||
0.022720 -0.044690 0.002354 -0.026181 -0.020685 0.025054 0.024201 0.023200 -0.039874 0.036060 -0.055478 -0.020999 0.029078 0.031083 0.001968 -0.055754 0.030704 0.028240 0.023472 -0.045850 0.061597 0.146746 0.052607 -0.080053 -0.001522 -0.164709 -0.062616 -0.095005 -0.127602 0.102249
|
||||
0.016286 -0.048030 -0.001184 -0.039975 -0.034766 0.032752 0.041787 0.015708 -0.046484 0.033420 -0.067307 -0.001994 0.020204 0.026865 -0.000861 -0.056410 0.032940 0.026198 -0.011653 -0.101627 0.003966 0.237505 0.036144 -0.055699 0.015305 -0.218927 -0.065851 -0.200727 -0.222999 0.153512
|
||||
0.075720 0.056380 -0.059414 0.024517 0.019087 0.077875 -0.100031 -0.030972 0.030338 -0.010825 0.016623 -0.143253 -0.030541 -0.005486 -0.117079 -0.030346 -0.013127 0.022452 0.072919 0.018036 -0.059039 -0.036724 -0.121192 0.015689 0.069917 -0.065978 -0.064873 -0.025976 0.031556 0.022245
|
||||
0.068039 0.034087 -0.043944 0.009759 0.026481 0.049916 -0.090858 -0.041117 0.025810 0.021895 -0.008543 -0.133490 -0.036091 0.013613 -0.091837 -0.035876 -0.001297 -0.009674 0.054274 -0.034062 -0.083163 -0.057757 -0.134940 -0.031745 0.021981 -0.068496 -0.003268 -0.028335 -0.015878 0.073877
|
||||
0.050150 0.006059 -0.024573 0.018377 0.027472 0.019785 -0.080324 -0.044483 0.022155 0.038587 -0.021428 -0.120383 -0.050280 0.040006 -0.088464 -0.045091 0.017150 -0.031191 0.019915 -0.051535 -0.079627 -0.096807 -0.081243 -0.091793 -0.117173 -0.034177 0.092763 -0.019522 -0.053013 0.134476
|
||||
0.033952 -0.019755 -0.012461 0.012433 0.027840 0.006878 -0.069979 -0.042903 0.015715 0.055837 -0.046298 -0.119158 -0.051559 0.049856 -0.046276 -0.044370 0.005964 -0.042713 -0.018983 -0.056325 -0.053581 -0.106128 -0.045745 -0.105085 -0.137515 -0.034382 0.129064 -0.002135 -0.096189 0.151989
|
||||
0.047280 0.009084 -0.026299 0.013747 0.030614 0.031466 -0.083539 -0.037043 0.029521 0.037780 -0.033683 -0.138624 -0.041837 0.036685 -0.080793 -0.056223 -0.013402 -0.018265 0.011874 -0.041360 -0.058506 -0.095337 -0.088379 -0.060889 -0.071366 -0.066558 0.059649 -0.006309 -0.044524 0.135719
|
||||
0.067275 0.041769 -0.048367 0.008016 0.030556 0.061809 -0.092514 -0.034013 0.027173 0.021460 -0.022642 -0.149936 -0.026244 0.012890 -0.090023 -0.041676 -0.033809 -0.000930 0.050481 -0.022578 -0.058594 -0.062894 -0.131553 -0.005734 0.052378 -0.089227 -0.020393 -0.016809 -0.013869 0.080783
|
||||
0.025142 0.001251 0.016565 -0.021801 -0.015814 -0.030528 0.026127 -0.047323 -0.014163 -0.009601 -0.067567 0.045374 0.092771 0.015750 -0.019568 -0.015963 -0.055046 0.098595 0.099420 -0.023551 -0.037654 -0.011608 -0.049887 0.236652 0.072817 0.011553 -0.152551 -0.017929 0.102193 -0.037352
|
||||
0.005739 -0.025054 0.028815 -0.028219 -0.013819 -0.046503 0.020039 -0.061044 -0.008280 0.001598 -0.074721 0.055569 0.073798 0.023639 0.015239 -0.008259 -0.061998 0.087132 0.084707 -0.067770 -0.040813 -0.037182 -0.072869 0.261121 0.015360 0.018577 -0.103545 -0.025366 0.068124 -0.011636
|
||||
-0.015805 -0.050892 0.047206 -0.018353 -0.016427 -0.075860 0.030657 -0.075106 -0.005075 0.014944 -0.091793 0.064329 0.058185 0.036443 0.018639 -0.003453 -0.039100 0.062342 0.047846 -0.084276 -0.028001 -0.075754 -0.018709 0.196795 -0.120455 0.034995 -0.004516 -0.026428 0.034617 0.023128
|
||||
-0.038583 -0.068385 0.061578 -0.031306 -0.017520 -0.100200 0.053688 -0.086437 -0.010096 0.045833 -0.112100 0.072391 0.063069 0.044706 0.057043 0.010307 -0.024131 0.023459 -0.000912 -0.097511 -0.018104 -0.090922 0.003572 0.110167 -0.152591 -0.001824 0.048390 -0.031118 -0.011626 0.024810
|
||||
-0.021322 -0.051141 0.049264 -0.016899 -0.027514 -0.089685 0.046146 -0.073606 -0.003933 0.031224 -0.099420 0.061318 0.054921 0.051073 -0.006545 -0.028813 -0.037431 0.060368 0.038592 -0.073295 -0.014447 -0.094122 -0.010330 0.193007 -0.138883 0.027769 -0.003818 -0.007939 0.030928 0.035641
|
||||
0.002855 -0.019624 0.028663 -0.023461 -0.024830 -0.060888 0.037252 -0.058740 -0.012505 0.018445 -0.081704 0.056135 0.070774 0.040713 -0.018203 -0.029022 -0.061647 0.080843 0.076472 -0.048515 -0.022364 -0.059220 -0.051196 0.252939 -0.022188 0.016877 -0.088539 -0.007235 0.061084 0.002476
|
||||
-0.010870 0.104712 0.029797 0.166451 0.076471 -0.016442 -0.046465 0.149026 0.086557 -0.039502 -0.076114 0.087570 -0.013519 -0.004196 0.034984 -0.007672 0.064149 0.048435 -0.064239 0.043959 0.009896 0.027999 -0.113348 -0.004341 -0.033293 0.065306 0.063698 -0.144347 0.103412 -0.052090
|
||||
0.017844 0.067857 0.033055 0.100714 0.011414 -0.014526 -0.033114 0.106862 0.028622 -0.032501 -0.064757 0.048714 -0.037980 0.016746 0.018282 0.009878 0.045060 -0.018655 -0.007535 0.079747 0.015679 0.072704 -0.095817 -0.016221 -0.065376 0.085619 -0.038141 -0.091654 0.048842 -0.070910
|
||||
0.030332 0.020687 0.013949 0.038061 0.013572 -0.000022 0.019596 0.081935 -0.013407 -0.005569 -0.045283 0.004474 -0.030739 0.033013 0.013673 0.045933 0.033542 -0.040870 0.004706 0.073059 0.017863 0.065842 -0.041730 -0.016454 -0.035901 0.063634 -0.059776 -0.067427 0.011136 -0.059024
|
||||
0.021858 -0.004289 0.000335 0.000132 0.001260 0.013361 0.038052 0.078177 -0.026188 0.004286 -0.020775 -0.007732 -0.015956 0.015308 0.018506 0.031945 0.022084 0.011279 -0.017030 0.020030 -0.026570 0.018856 0.004329 -0.031829 -0.007195 -0.001821 0.001326 -0.039582 0.025319 -0.033323
|
||||
0.006263 -0.031056 -0.012718 -0.039882 -0.018196 0.027894 0.045726 0.073221 -0.024310 0.025043 0.020388 -0.022822 -0.017723 -0.011411 0.023485 0.008443 0.025824 0.075306 -0.005381 -0.028184 -0.065877 -0.014889 0.056643 -0.045479 0.017933 -0.061061 0.081802 0.003605 0.039520 -0.034431
|
||||
-0.028803 -0.044499 -0.041397 -0.099367 -0.020188 0.038397 0.082539 0.061655 -0.046445 0.030438 0.054320 -0.049559 -0.018808 -0.037850 0.022436 0.035089 0.010324 0.017405 -0.002013 -0.067517 -0.094251 -0.038230 0.050493 0.019274 0.032527 -0.068609 0.122562 -0.002028 0.019055 -0.069501
|
||||
-0.065346 -0.030968 -0.056167 -0.162442 -0.087023 0.032999 0.092738 0.054231 -0.092953 0.000349 0.045766 -0.069881 -0.068998 -0.076344 -0.006702 0.064521 -0.013521 -0.110076 0.009722 -0.071902 -0.148699 -0.023199 -0.006040 0.094013 0.008338 -0.027304 0.115529 -0.015983 -0.013156 -0.135132
|
||||
-0.034679 -0.043523 -0.052291 -0.113822 -0.041847 0.033727 0.099498 0.071348 -0.052046 0.002160 0.018334 -0.034949 -0.022358 -0.056131 0.003802 0.035901 -0.002773 0.023212 -0.080146 -0.058363 -0.159504 -0.044258 0.049508 -0.019479 0.030311 -0.042848 0.040935 0.037282 -0.012424 -0.011143
|
||||
-0.001188 -0.029947 -0.021378 -0.050264 -0.030362 0.027528 0.074933 0.086762 -0.024414 -0.003290 -0.016840 -0.006605 -0.011018 -0.023567 0.013467 0.010798 -0.011063 0.095323 -0.143006 -0.016337 -0.151915 -0.058152 0.095316 -0.054460 0.058470 -0.041779 -0.029734 0.073772 0.014706 0.112921
|
||||
0.018355 0.004699 0.002318 -0.001406 -0.014467 0.008747 0.044992 0.098434 -0.018868 -0.002335 -0.035804 -0.007438 -0.029446 0.003799 0.018935 0.034637 -0.007953 0.020178 -0.123839 0.076915 -0.081495 -0.058879 0.032391 0.027460 0.044353 0.047615 -0.077064 0.064515 -0.033796 0.100925
|
||||
0.033019 0.043835 0.023435 0.044337 -0.001423 -0.011391 0.004388 0.114947 -0.002926 0.000787 -0.038694 -0.007835 -0.062049 0.021805 0.022203 0.051078 0.006535 -0.047992 -0.065916 0.162790 -0.008000 -0.043768 -0.026127 0.112196 0.021601 0.137598 -0.093711 0.055645 -0.082083 0.051915
|
||||
0.017273 0.084271 0.043289 0.114244 0.018058 -0.016534 -0.041516 0.135901 0.035920 -0.026585 -0.051314 0.042434 -0.054356 0.009469 0.039614 0.025958 0.031490 -0.026176 -0.044059 0.120278 0.010208 -0.004451 -0.068608 0.090021 -0.005751 0.114878 -0.019074 -0.038092 0.008491 0.008537
|
||||
-0.016626 0.103489 0.036047 0.170977 0.070786 -0.011811 -0.052227 0.158526 0.083324 -0.037709 -0.061632 0.110622 -0.040038 -0.003881 0.016006 -0.042596 0.079501 0.068068 -0.045883 0.078107 -0.000844 0.010056 -0.127653 0.007866 -0.030414 0.027243 0.095718 -0.117872 0.088501 -0.058162
|
||||
0.034671 0.016690 0.018371 0.038280 0.018539 -0.004089 0.012754 0.086368 -0.009429 0.003495 -0.040590 -0.003348 -0.039763 0.035331 0.008499 0.046292 0.029303 -0.044206 -0.026758 0.127168 0.018570 0.026200 -0.061873 -0.000662 -0.017469 0.071814 -0.064138 -0.056290 -0.028169 -0.050189
|
||||
0.027182 -0.005360 0.001933 -0.002120 -0.000599 0.009605 0.039104 0.080512 -0.027057 0.004084 -0.023116 -0.008426 -0.017939 0.012235 0.016945 0.036639 0.011825 0.014256 -0.053941 0.040437 -0.040286 -0.001633 0.004695 -0.019027 0.004648 -0.001881 -0.011982 -0.018817 0.009917 -0.001713
|
||||
0.013662 -0.030688 -0.013993 -0.045674 -0.023338 0.023408 0.050240 0.072713 -0.029837 0.012841 0.007414 -0.020122 -0.015171 -0.024405 0.021587 0.021321 0.007006 0.084412 -0.048291 -0.049120 -0.097299 -0.021608 0.069445 -0.035270 0.021177 -0.063787 0.067497 0.031093 0.039775 0.024742
|
||||
-0.063276 -0.025452 -0.062397 -0.165996 -0.078886 0.030190 0.102017 0.062863 -0.093828 0.000325 0.035281 -0.089094 -0.047858 -0.081959 0.019795 0.104221 -0.027445 -0.130634 -0.013468 -0.106960 -0.136921 -0.001032 -0.004673 0.081097 0.001390 -0.003926 0.096202 -0.050716 0.004331 -0.141428
|
||||
0.009040 -0.030850 -0.018624 -0.052646 -0.025465 0.024925 0.059354 0.078963 -0.023758 0.002993 -0.004478 -0.015417 -0.020069 -0.039726 0.024858 0.033557 0.005765 0.100249 -0.090534 -0.037587 -0.141149 -0.045093 0.090303 -0.044298 0.039907 -0.042388 0.028304 0.046166 0.033213 0.077550
|
||||
0.024602 -0.001576 0.003707 -0.001173 -0.008097 0.009008 0.038778 0.088839 -0.024916 -0.003596 -0.028066 -0.004878 -0.029700 0.001871 0.023320 0.051616 0.010695 0.021845 -0.087960 0.058211 -0.073707 -0.026869 0.019480 -0.002572 0.027466 0.025521 -0.038530 0.015489 -0.018891 0.034746
|
||||
0.033851 0.026804 0.025613 0.047635 0.006103 -0.006719 0.004629 0.097295 -0.013383 -0.001173 -0.042797 -0.000701 -0.056687 0.031313 0.014096 0.062815 0.029375 -0.044625 -0.051873 0.154272 -0.008023 -0.002000 -0.049674 0.047529 0.008205 0.097629 -0.078808 -0.005855 -0.070467 -0.031403
|
||||
-0.047961 0.093209 0.003265 -0.046195 0.129364 -0.102661 -0.107139 -0.077072 -0.207712 0.043826 0.041516 0.002694 -0.101096 -0.143447 -0.199436 0.280060 -0.059966 0.222904 0.176204 -0.061353 0.094202 0.015620 -0.091721 0.014665 0.003234 0.037080 0.070471 0.060765 0.068863 0.223408
|
||||
0.015591 0.066202 0.000466 -0.055129 0.104971 -0.117132 -0.145469 -0.097254 -0.124733 0.065984 0.040786 0.001346 -0.073768 -0.084814 -0.105047 0.205150 0.021672 0.177949 0.122439 -0.035318 0.029959 -0.014803 -0.016775 0.011985 0.022156 0.014810 0.016482 0.022173 -0.007147 0.072586
|
||||
0.075248 0.038299 -0.006472 -0.057770 0.079234 -0.117493 -0.173452 -0.124616 -0.065209 0.085805 0.043464 0.013280 -0.030767 -0.020266 -0.006488 0.119855 0.093738 0.109988 0.055993 0.005578 -0.040280 -0.030758 0.054947 -0.001917 0.046533 -0.024006 -0.038891 -0.027673 -0.036479 -0.065925
|
||||
0.120483 0.012097 -0.016978 -0.060845 0.057695 -0.093498 -0.190101 -0.143898 -0.024796 0.083386 0.049099 0.014165 0.013479 0.020865 0.071114 0.035555 0.135863 0.036062 -0.015163 0.037736 -0.101069 -0.044151 0.100665 -0.003043 0.029146 -0.052474 -0.070448 -0.078753 -0.026905 -0.161501
|
||||
0.139933 -0.009970 -0.033370 -0.060182 0.044890 -0.034069 -0.196608 -0.140680 0.000853 0.052715 0.049924 0.005022 0.048856 0.047123 0.108746 -0.036667 0.128729 -0.021106 -0.075016 0.061149 -0.114590 -0.077483 0.086677 0.039284 -0.008786 -0.048348 -0.050485 -0.111390 -0.007411 -0.159964
|
||||
0.127349 -0.030754 -0.063498 -0.046493 0.034211 0.046070 -0.183837 -0.122364 0.010019 0.006563 0.043433 0.006138 0.073516 0.082966 0.113144 -0.078262 0.066061 -0.032655 -0.120280 0.088806 -0.077527 -0.091363 0.034158 0.097651 -0.032819 -0.054274 -0.006854 -0.089536 0.015376 -0.086009
|
||||
0.091647 -0.051086 -0.109411 -0.021640 0.016727 0.125047 -0.137158 -0.097738 0.008104 -0.050609 0.030280 0.028727 0.075621 0.142032 0.081896 -0.072985 -0.016124 0.001182 -0.137944 0.109089 -0.015088 -0.045163 -0.026852 0.123807 -0.039824 -0.071373 0.047268 -0.013109 0.039526 0.026208
|
||||
0.037734 -0.068199 -0.152408 0.002743 -0.005221 0.184749 -0.056023 -0.058803 0.000079 -0.100872 -0.009278 0.050239 0.023925 0.190397 0.022450 -0.014604 -0.061295 0.061675 -0.099756 0.115791 0.065619 0.074478 -0.086820 0.079713 -0.042578 -0.065624 0.115590 0.090235 0.072581 0.114556
|
||||
0.005335 -0.065374 -0.178339 0.021993 -0.009750 0.219907 0.007917 -0.025263 -0.026174 -0.151432 -0.068636 0.068063 -0.036603 0.204249 -0.010333 0.077472 -0.058110 0.114683 -0.021601 0.100297 0.094381 0.158266 -0.071317 0.001749 -0.022941 -0.052596 0.154666 0.082033 0.088986 0.101366
|
||||
-0.017694 -0.020177 -0.173228 0.015658 -0.020889 0.201771 -0.013394 0.041426 -0.003011 -0.106398 -0.096898 0.025251 -0.060957 0.137257 0.040566 0.087226 0.016338 0.124550 0.070381 0.032883 0.090008 0.125352 -0.032893 -0.001077 0.010035 0.002425 0.173421 0.073592 0.008836 -0.012713
|
||||
-0.045583 0.065863 -0.150586 -0.004698 -0.019784 0.147752 -0.039052 0.122879 0.011238 -0.067095 -0.098236 -0.004236 -0.046488 0.064507 0.102239 0.052923 0.062526 0.069713 0.134455 -0.065098 0.097477 0.018140 0.050997 0.016748 0.031419 0.065554 0.069262 0.049984 -0.107316 -0.056271
|
||||
-0.064791 0.130743 -0.114224 -0.032004 -0.014633 0.063912 -0.045598 0.177868 0.010467 -0.020958 -0.062875 -0.013081 -0.024540 0.022590 0.115642 -0.006494 0.090701 -0.008540 0.136552 -0.119375 0.087294 -0.047868 0.085050 0.032881 0.002264 0.093251 -0.051055 0.039511 -0.129941 -0.026233
|
||||
-0.069557 0.160593 -0.080864 -0.052313 -0.007646 -0.023805 -0.031650 0.215543 -0.013424 0.017763 -0.017771 0.005663 -0.003392 0.026401 0.088687 -0.053326 0.093736 -0.072587 0.094140 -0.133131 0.054553 -0.054082 0.070270 0.039797 -0.037413 0.067557 -0.139764 0.037952 -0.103414 0.038027
|
||||
-0.053517 0.147594 -0.048502 -0.063337 0.008996 -0.093419 -0.023693 0.233132 -0.060622 0.044604 0.028397 0.033322 0.022070 0.054132 0.047446 -0.069260 0.049913 -0.082495 0.034166 -0.103206 0.019681 -0.028623 0.027775 0.022442 -0.051774 0.027490 -0.149326 0.029956 -0.068210 0.055083
|
||||
-0.020653 0.103733 -0.010655 -0.071903 0.040518 -0.130522 -0.033209 0.224597 -0.113374 0.050689 0.066346 0.043502 0.036927 0.060469 -0.019912 -0.069852 -0.031983 -0.045716 -0.045328 -0.021897 0.023060 -0.018184 -0.006866 -0.001889 -0.035069 -0.007022 -0.078842 0.011285 -0.051878 -0.001704
|
||||
0.021280 0.045013 0.026989 -0.083221 0.079652 -0.143603 -0.044720 0.191758 -0.177704 0.036307 0.086946 0.043019 0.038864 0.042521 -0.110858 -0.068091 -0.130767 0.001664 -0.124953 0.077152 0.046989 -0.008634 -0.015449 -0.017420 -0.011248 -0.038139 0.027524 0.012795 -0.024354 -0.063614
|
||||
0.065635 -0.018488 0.061515 -0.088365 0.119386 -0.141281 -0.048132 0.152667 -0.265313 0.018702 0.095882 0.049978 0.047730 0.019882 -0.199261 -0.050768 -0.225388 0.044395 -0.189686 0.176789 0.068785 0.017459 -0.021862 -0.048674 0.019516 -0.083740 0.137119 0.003448 0.025716 -0.098104
|
||||
-0.047457 0.004826 0.086340 -0.136927 0.003215 0.088356 -0.091942 0.054053 0.300055 0.126842 -0.093633 -0.266702 0.014134 0.109025 -0.145511 -0.094568 -0.133065 0.092700 -0.043392 -0.127601 0.023627 0.124558 0.070196 0.073132 -0.088606 0.089546 -0.033665 -0.056504 0.100204 -0.082697
|
||||
-0.043093 0.054162 0.063821 -0.126107 -0.014163 0.082007 -0.031381 0.031526 0.175988 0.115797 -0.135266 -0.115958 0.154077 0.045811 -0.107576 -0.020057 -0.105244 -0.000702 -0.047848 -0.077532 0.078558 0.088457 -0.008710 0.022423 -0.020940 -0.003605 0.016126 0.001175 0.058905 -0.128046
|
||||
-0.049606 0.069200 0.053290 -0.109077 -0.046531 0.083786 0.018246 0.007123 0.096077 0.119081 -0.133557 0.051326 0.172765 0.028501 -0.086385 -0.005934 -0.051025 -0.069052 -0.012008 -0.027332 0.071679 0.012252 -0.050119 -0.015617 0.032504 -0.045614 0.060595 0.018311 0.002522 -0.129035
|
||||
-0.050327 0.056497 0.052203 -0.091244 -0.070921 0.096400 0.053644 -0.015881 0.055839 0.109643 -0.104617 0.172271 0.109450 0.053063 -0.094297 -0.016023 -0.008077 -0.106474 0.037792 0.020494 0.034454 -0.069321 -0.043290 -0.035857 0.083344 -0.045902 0.084348 -0.008476 -0.015850 -0.075485
|
||||
-0.045033 0.029609 0.054298 -0.078579 -0.083367 0.108093 0.069142 -0.033561 0.037172 0.099455 -0.048139 0.255389 -0.018561 0.095840 -0.124772 -0.039076 0.018613 -0.124882 0.104486 0.078116 0.002609 -0.162990 -0.015797 -0.015448 0.134282 -0.020490 0.100475 -0.044110 -0.021376 0.012862
|
||||
0.018337 -0.024647 0.071908 -0.080413 -0.061404 0.121688 0.061343 -0.015781 -0.002231 0.054575 -0.034324 0.267931 -0.023291 0.118740 -0.092321 0.054178 0.078846 -0.085469 0.136116 0.074133 -0.108748 -0.179256 0.011133 -0.049144 0.075315 -0.007033 0.020333 -0.049683 -0.012017 0.130313
|
||||
0.033647 -0.017865 0.077483 -0.095394 -0.031197 0.118052 0.019863 0.003291 0.013741 0.033189 -0.042590 0.218113 0.082374 0.057508 -0.028366 0.110658 0.081760 -0.082289 0.059756 0.026578 -0.129004 -0.052370 0.011842 -0.062670 0.042701 0.046589 -0.003135 0.011039 -0.018942 0.088763
|
||||
0.037987 -0.013052 0.079243 -0.111549 0.003707 0.116562 -0.038333 0.029299 0.058843 0.010267 -0.011673 0.140940 0.114290 -0.009354 0.016953 0.164719 0.076837 -0.070848 -0.010215 -0.004144 -0.104697 0.048368 0.011353 -0.033681 -0.005250 0.075743 -0.024358 0.069399 -0.015366 0.052231
|
||||
0.024571 -0.014004 0.081088 -0.124925 0.039212 0.120645 -0.104834 0.052231 0.161371 -0.029810 0.055543 0.031504 0.051553 -0.080459 0.034547 0.214264 0.064089 -0.057701 -0.093159 -0.032200 -0.032440 0.114309 0.033187 0.029077 -0.074945 0.029941 -0.063097 0.100517 0.012563 0.045334
|
||||
-0.002837 -0.038231 0.092274 -0.131858 0.049030 0.126598 -0.163394 0.063928 0.305740 -0.046989 0.136958 -0.067484 -0.123285 -0.115755 0.023003 0.209179 0.061858 -0.027002 -0.172347 -0.047980 0.050874 0.104938 0.077753 0.093367 -0.173518 -0.104694 -0.132603 0.087518 0.014903 0.062252
|
||||
0.035278 0.093079 0.010600 0.011548 -0.002504 -0.023582 0.180448 -0.078628 -0.088358 -0.202506 0.026650 0.100765 -0.085380 0.151737 -0.205276 0.021165 0.026303 0.014300 -0.001564 0.011926 -0.007819 0.011798 0.298873 0.061428 -0.063378 -0.105397 -0.060695 -0.098906 0.122579 -0.008992
|
||||
0.017390 0.050818 0.013606 0.009554 -0.068511 -0.042936 0.116296 -0.057306 -0.074851 -0.123759 0.038330 0.038347 -0.067773 0.114332 -0.168521 -0.024855 0.124354 0.058855 -0.047706 -0.014747 0.036327 0.017807 0.148729 0.075751 -0.037770 -0.032198 -0.027718 -0.088111 -0.005536 -0.011614
|
||||
0.001168 0.012468 0.014343 0.005583 -0.136504 -0.058055 0.047533 -0.031252 -0.059431 -0.043098 0.054520 -0.022029 -0.047399 0.069219 -0.133415 -0.070029 0.209996 0.108755 -0.081356 -0.035798 0.082904 0.018707 -0.009269 0.100721 0.000104 0.044993 0.005827 -0.068919 -0.125661 -0.017308
|
||||
-0.013446 -0.023076 0.014190 0.001928 -0.202312 -0.067740 -0.022495 -0.003279 -0.040029 0.038724 0.069213 -0.077248 -0.031443 0.028175 -0.092024 -0.112996 0.287767 0.157286 -0.112861 -0.058615 0.128308 0.029739 -0.171848 0.132390 0.037112 0.116335 0.032699 -0.048071 -0.248989 -0.017920
|
||||
-0.004607 -0.002814 0.034971 0.063651 -0.079836 -0.058991 -0.000815 -0.009434 -0.063895 0.008389 -0.015031 -0.095108 -0.011112 0.018624 0.006932 -0.007569 0.083353 -0.063261 -0.116140 -0.083180 0.064087 -0.108477 0.007666 -0.028230 0.104627 0.051982 -0.021014 0.111445 0.196977 -0.008409
|
||||
-0.018307 -0.033552 0.029480 0.053194 -0.113895 -0.039754 -0.008420 -0.018112 -0.076130 0.009463 -0.016153 -0.081128 -0.021176 0.028977 0.002168 -0.009029 0.118326 -0.018611 -0.135946 -0.117746 0.077541 -0.042099 -0.069344 -0.050713 0.091173 0.039218 -0.023106 0.006241 0.112340 0.031851
|
||||
-0.031895 -0.072550 0.031542 0.045525 -0.143291 -0.022734 -0.016811 -0.028304 -0.076746 0.021967 0.000082 -0.075735 -0.042067 0.038968 0.005811 -0.019303 0.156584 0.026586 -0.142034 -0.128677 0.090010 -0.004233 -0.127694 -0.078041 0.087218 0.040231 -0.016823 -0.040045 0.073646 0.039582
|
||||
-0.014161 -0.038147 0.030604 0.054953 -0.110218 -0.038330 -0.013869 -0.026526 -0.074202 -0.003619 0.004746 -0.067246 -0.022840 0.032457 0.009291 0.004525 0.119326 -0.020736 -0.110889 -0.142146 0.072134 -0.026608 -0.087958 -0.063928 0.074964 0.051145 -0.002037 -0.026674 0.118991 -0.010838
|
||||
-0.004696 -0.012051 0.035134 0.069484 -0.075314 -0.054016 -0.011467 -0.025499 -0.058403 -0.018792 0.026287 -0.062437 -0.033176 0.025773 0.018710 0.024651 0.088722 -0.062940 -0.048949 -0.147240 0.052459 -0.078277 -0.052000 -0.039959 0.056470 0.066799 0.028366 0.029261 0.218751 -0.099509
|
||||
-0.049581 0.033585 0.057599 -0.052382 0.044515 -0.018327 0.074956 -0.038808 0.008300 -0.013971 -0.093913 -0.123879 -0.001633 -0.052343 0.102658 -0.132508 -0.089045 0.020652 -0.033426 0.034692 0.088334 -0.021309 -0.020739 -0.001398 0.004088 -0.047299 0.025109 0.056241 -0.015862 0.039179
|
||||
-0.032619 0.050779 0.046419 -0.065647 0.037767 0.008875 0.098799 -0.042597 -0.005152 -0.052716 -0.079103 -0.080974 0.002141 -0.051647 0.146710 -0.167109 -0.114777 0.061824 0.038295 0.058887 0.038151 0.054364 -0.125482 0.061687 0.219252 -0.129146 -0.112676 0.048264 -0.018227 -0.042494
|
||||
-0.031082 0.042306 0.043899 -0.058791 0.031868 0.009922 0.112119 -0.050662 -0.021540 -0.081705 -0.053208 -0.051498 -0.023552 -0.036828 0.128762 -0.150411 -0.094850 0.057711 0.047094 0.062646 0.019487 0.045503 -0.111206 0.049679 0.194214 -0.119797 -0.107132 0.026785 -0.005012 -0.040908
|
||||
-0.022489 0.025532 0.055078 -0.035242 0.030775 -0.036223 0.103658 -0.056472 -0.018920 -0.069823 -0.032603 -0.061203 -0.064638 -0.008579 0.031144 -0.098991 -0.011993 -0.020266 -0.015539 0.021713 0.027393 -0.059495 0.061798 -0.026476 -0.095092 -0.007043 0.055712 0.014933 -0.005036 0.060440
|
||||
-0.030310 0.033513 0.064336 -0.018414 0.034721 -0.048560 0.109454 -0.053871 -0.016464 -0.078713 -0.072262 -0.067361 -0.045979 -0.023773 0.069114 -0.121990 -0.037716 -0.007418 -0.016633 0.044367 0.052566 -0.057098 0.060782 -0.051123 -0.123901 0.021446 0.071002 0.033668 -0.035860 0.069226
|
||||
-0.043942 0.026216 0.071600 -0.023807 0.037164 -0.046979 0.093011 -0.047115 -0.003050 -0.042657 -0.087662 -0.094068 -0.029746 -0.036867 0.097107 -0.136982 -0.053572 0.001892 -0.027773 0.053083 0.072788 -0.056732 0.024396 -0.049643 -0.104153 0.006579 0.080705 0.050583 -0.054139 0.067745
|
||||
0.006750 -0.003083 0.064063 -0.036630 0.045616 -0.026923 0.081208 -0.053243 -0.029335 -0.109279 0.018407 -0.023073 -0.083213 -0.020485 0.066702 -0.033207 0.038323 -0.017964 0.018759 -0.001107 -0.045386 -0.031948 0.051387 -0.053143 -0.119762 0.048682 0.076599 -0.028281 0.010670 0.037614
|
||||
0.021021 -0.006153 0.059366 -0.062042 0.056695 0.023990 0.076039 -0.048741 -0.041687 -0.139957 0.017771 0.003256 -0.043052 -0.050945 0.186826 -0.055048 -0.016978 0.066238 0.097825 0.029590 -0.095129 0.080565 -0.116792 0.024034 0.151924 -0.042052 -0.092912 -0.016833 0.016146 -0.039069
|
||||
0.023721 -0.009506 0.064424 -0.065985 0.070977 0.030273 0.048462 -0.037688 -0.026275 -0.133367 0.033100 0.000599 -0.041355 -0.074379 0.225882 -0.040265 -0.011023 0.063907 0.100378 0.018538 -0.110506 0.098623 -0.148510 0.028398 0.152254 -0.027240 -0.092449 -0.015261 0.009466 -0.053450
|
||||
0.019763 -0.036128 0.079128 -0.053098 0.082815 0.005005 0.017600 -0.035142 -0.014838 -0.108518 0.034834 -0.028918 -0.050576 -0.078779 0.193010 0.006929 0.028337 0.015610 0.040762 -0.012010 -0.081863 0.027171 -0.050422 -0.046367 -0.066804 0.074067 0.053057 -0.029078 0.015778 0.002879
|
||||
0.011938 -0.029503 0.087532 -0.024420 0.069483 -0.027846 0.044993 -0.047368 -0.020828 -0.118975 0.010482 -0.020697 -0.063108 -0.057586 0.166967 -0.021956 0.037222 -0.000269 0.037193 0.000745 -0.066657 -0.017101 -0.011113 -0.079866 -0.162095 0.101422 0.113248 -0.028325 -0.021407 0.036151
|
||||
0.013043 -0.009330 0.078092 -0.019995 0.058841 -0.036270 0.072063 -0.054900 -0.027719 -0.133535 0.003695 -0.011225 -0.073466 -0.038105 0.120544 -0.033702 0.031235 -0.005709 0.029226 0.006591 -0.056758 -0.025869 0.025826 -0.065962 -0.171734 0.088452 0.102577 -0.031146 -0.007347 0.046178
|
||||
-0.007658 0.004009 0.025908 0.154403 0.053330 -0.011857 -0.013594 0.056724 0.061838 0.101450 0.018192 -0.052999 0.071364 -0.068356 0.093998 0.018847 -0.115542 -0.055564 -0.072179 0.036748 0.001687 -0.144735 0.108376 0.148961 0.128366 -0.004208 0.109635 0.032357 -0.050662 0.200893
|
||||
-0.002996 0.005226 0.032797 0.135787 -0.089440 -0.040576 -0.057551 0.032946 0.001834 0.032712 -0.025593 -0.049533 0.022227 -0.047612 0.014361 0.009649 -0.131089 -0.094849 -0.020649 0.086202 -0.039909 -0.001707 0.056046 0.007381 0.033777 -0.002541 -0.034926 -0.004427 -0.026112 0.123944
|
||||
-0.004291 -0.005543 0.022541 0.113681 -0.173953 -0.041398 -0.084376 0.020405 -0.021779 -0.018671 -0.025165 -0.035068 -0.015855 -0.040666 -0.039493 0.008537 -0.120702 -0.084443 0.035688 0.098020 -0.048034 0.087448 -0.002553 -0.045851 -0.023169 0.006651 -0.103322 -0.059515 -0.002982 0.053502
|
||||
-0.004726 -0.017325 0.020839 0.120502 -0.184684 -0.030813 -0.080318 0.007868 -0.032129 -0.044198 -0.019700 -0.008229 -0.027685 -0.022684 -0.063155 0.021599 -0.119028 -0.076535 0.072624 0.078086 -0.069180 0.100850 0.025706 -0.061785 -0.040139 -0.018370 -0.106073 -0.066669 0.007369 0.027967
|
||||
-0.000280 -0.015779 0.023768 0.114944 -0.170625 -0.036464 -0.096963 0.008030 -0.019057 -0.023453 -0.012107 -0.028049 -0.023436 -0.043491 -0.030102 0.016863 -0.097956 -0.080536 0.095216 0.070681 -0.042470 0.108651 -0.007977 -0.072278 -0.048337 0.008349 -0.078533 -0.092741 0.008143 -0.000355
|
||||
0.001835 -0.014562 0.035395 0.138589 -0.082041 -0.030999 -0.083040 0.009139 0.007790 0.017249 0.017982 -0.022819 0.007465 -0.056808 0.031589 0.035086 -0.080570 -0.087719 0.127579 0.004619 -0.028175 0.053466 0.013858 -0.040136 -0.048417 -0.015858 0.045949 -0.125270 0.015984 -0.035032
|
||||
0.006921 -0.025647 0.033160 0.154624 0.058888 -0.002069 -0.036401 0.017355 0.058515 0.088564 0.060386 -0.030704 0.068512 -0.065203 0.113409 0.060767 -0.056340 -0.035930 0.137062 -0.085671 0.034591 -0.049785 0.047278 0.057771 0.011576 -0.039723 0.207910 -0.149281 0.010540 -0.014703
|
||||
-0.018065 -0.076293 -0.069286 0.078198 0.101563 -0.013276 0.078677 0.006820 0.089090 0.096060 0.070680 0.003228 0.021392 -0.088395 -0.008440 0.012816 0.036751 -0.005407 0.105001 -0.058448 -0.008691 0.008147 -0.062581 -0.013663 -0.046333 0.018932 0.094530 -0.020238 -0.044608 -0.178033
|
||||
-0.023295 -0.105717 -0.130887 0.042172 0.126279 -0.022736 0.153202 -0.009995 0.089795 0.083969 0.040114 0.033834 0.021716 -0.078432 -0.083662 -0.014406 0.064186 0.025477 0.011953 -0.052429 -0.063783 0.018598 -0.086487 -0.094668 -0.053926 0.006797 -0.024294 0.066220 -0.053567 -0.167064
|
||||
-0.018044 -0.105365 -0.143033 0.038317 0.122695 -0.024418 0.180070 -0.016368 0.080389 0.066925 0.023230 0.050475 0.022090 -0.061951 -0.122699 -0.026327 0.044794 0.048222 -0.037006 -0.034314 -0.083880 0.005961 -0.055945 -0.101092 -0.035779 -0.012444 -0.081835 0.102054 -0.043497 -0.108638
|
||||
-0.014769 -0.102335 -0.129133 0.038868 0.120484 -0.030852 0.167176 -0.010452 0.086453 0.097577 0.016205 0.019218 0.031039 -0.063618 -0.104403 -0.032292 0.029241 0.036091 -0.070427 0.000187 -0.069704 -0.024270 -0.056930 -0.054254 -0.012952 0.009929 -0.089541 0.151599 -0.081143 -0.050335
|
||||
-0.009481 -0.070383 -0.068210 0.074523 0.094020 -0.024829 0.096223 0.012112 0.085339 0.121468 0.024542 -0.025869 0.034262 -0.065157 -0.033181 -0.027444 -0.026184 0.003234 -0.078000 0.059224 -0.036972 -0.092057 -0.010613 0.070660 0.038012 0.033748 -0.040853 0.163250 -0.097112 0.068403
|
||||
-0.006421 -0.000488 0.015258 0.144050 0.033823 -0.016970 -0.006361 0.047553 0.045559 0.085494 0.014984 -0.044455 0.059172 -0.064238 0.064522 0.014732 -0.103781 -0.058687 -0.051086 0.036277 -0.013219 -0.110463 0.090567 0.119696 0.104685 -0.002963 0.082936 0.046848 -0.050252 0.162079
|
||||
-0.002256 -0.003070 0.031345 0.153930 -0.123343 -0.028048 -0.036352 0.000850 -0.045851 -0.025909 -0.016508 -0.019979 -0.013828 -0.019081 -0.050397 0.011483 -0.100915 -0.088543 0.022607 -0.001025 -0.117985 0.063391 0.067644 0.038103 0.006033 -0.085097 -0.039441 0.146388 -0.041772 -0.051340
|
||||
0.000239 -0.007604 0.023409 0.154526 -0.136859 -0.023180 -0.025502 -0.012380 -0.055244 -0.058817 -0.011823 0.005713 -0.030494 -0.001273 -0.075493 0.026846 -0.099485 -0.082640 0.049857 -0.020651 -0.130935 0.088740 0.090593 0.008703 -0.018623 -0.108109 -0.053272 0.124024 -0.022554 -0.081954
|
||||
0.001199 -0.011065 0.032638 0.154798 -0.119491 -0.024075 -0.045495 -0.006920 -0.045810 -0.032928 -0.000337 -0.011610 -0.020302 -0.023645 -0.043698 0.030470 -0.082403 -0.082719 0.076415 -0.033587 -0.114253 0.087562 0.055641 0.003736 -0.017577 -0.091832 -0.005653 0.106889 -0.033713 -0.097238
|
||||
0.005941 -0.025372 0.021699 0.143992 0.039195 -0.009543 -0.024260 0.012936 0.042846 0.075542 0.052968 -0.023932 0.056747 -0.061245 0.077369 0.051064 -0.054227 -0.041527 0.122572 -0.061339 0.015604 -0.031794 0.034546 0.045040 0.004776 -0.039600 0.169167 -0.117533 0.005141 -0.028689
|
||||
-0.007357 -0.075194 -0.139004 0.012430 0.082505 -0.022773 0.116595 0.009288 0.080244 0.046904 0.034180 0.024875 0.019728 -0.103957 -0.069610 0.026953 -0.042162 0.047266 0.035200 0.018932 0.033693 0.012070 -0.058543 -0.120564 -0.046493 0.092326 -0.096271 -0.188280 0.093647 0.032039
|
||||
-0.002783 -0.073032 -0.152808 0.011773 0.075152 -0.022645 0.139977 0.002972 0.078749 0.025628 0.022546 0.044844 0.011951 -0.081888 -0.106012 0.013315 -0.061439 0.061630 -0.001427 0.041373 0.010233 0.008815 -0.021219 -0.121332 -0.039220 0.075509 -0.147920 -0.169837 0.109254 0.075214
|
||||
-0.003319 -0.072659 -0.138051 0.010696 0.078350 -0.027187 0.124494 0.007657 0.079696 0.059854 0.015961 0.011495 0.027594 -0.090032 -0.079870 0.004849 -0.063277 0.046793 -0.033500 0.067179 0.022644 -0.021069 -0.042023 -0.085950 -0.011708 0.102277 -0.147686 -0.126063 0.083253 0.113313
|
||||
0.213194 0.016902 0.118495 -0.032098 -0.070767 -0.100405 0.132766 0.009845 0.060037 -0.086622 -0.235538 0.028654 -0.143848 -0.001808 -0.040712 0.204901 -0.125084 -0.041909 -0.094990 -0.087732 0.088228 -0.002044 -0.118188 0.030456 -0.031721 0.036866 0.070312 -0.032136 -0.085564 -0.007507
|
||||
0.223609 -0.009617 0.159810 -0.030141 -0.084089 -0.041429 0.096320 0.022277 0.062519 -0.019070 -0.108678 0.004209 -0.037498 -0.025240 -0.014830 0.137475 -0.058222 0.081868 -0.016233 0.012772 0.050980 -0.019823 -0.048442 -0.015028 -0.004558 0.035297 -0.001270 0.001153 -0.050428 0.026409
|
||||
0.218117 -0.040337 0.195755 -0.015467 -0.088445 0.026181 0.045735 0.034464 0.051706 0.051759 0.032196 -0.005602 0.051044 -0.044612 0.005116 0.036107 0.014768 0.179215 0.073844 0.086469 0.024887 -0.032072 0.010939 -0.040091 0.035717 0.005682 -0.063479 0.021971 -0.036761 0.041887
|
||||
0.162908 -0.066824 0.201691 -0.010338 -0.062296 0.090158 -0.001047 0.073741 0.004624 0.094418 0.148675 -0.009344 0.079161 -0.059963 0.000962 -0.081890 0.074739 0.166268 0.097326 0.114165 0.021648 -0.006801 0.035708 -0.037302 0.022061 -0.021397 -0.049087 0.033870 -0.009254 0.023934
|
||||
0.081992 -0.072911 0.180235 -0.024173 0.006782 0.122361 -0.024445 0.125018 -0.066064 0.071804 0.152290 -0.025533 0.063364 -0.055950 -0.020966 -0.123691 0.078293 0.051322 0.037715 0.060414 0.025468 0.032247 0.008974 0.006349 -0.012643 -0.028394 0.027379 -0.012245 0.017741 0.007820
|
||||
0.020482 -0.063116 0.138540 -0.034154 0.084272 0.087832 -0.038205 0.144056 -0.116625 0.032919 0.059064 -0.034380 0.034258 -0.029512 -0.051806 -0.092486 0.056760 -0.054640 -0.037899 -0.005119 -0.008487 0.068641 -0.016545 0.031151 -0.017460 -0.056197 0.093089 -0.030210 0.058728 -0.012908
|
||||
0.001980 -0.021483 0.086910 -0.018888 0.163524 0.013444 -0.032083 0.137051 -0.118495 0.001522 -0.092606 -0.048554 -0.002407 0.009327 -0.073447 0.000238 0.031210 -0.108458 -0.063692 -0.074288 -0.054189 0.077708 0.004916 0.046374 0.018212 -0.071675 0.101690 0.013387 0.052165 -0.055058
|
||||
0.013546 0.004061 0.043329 0.008350 0.245832 -0.040721 -0.053514 0.061704 -0.083365 -0.033573 -0.170121 -0.048003 -0.026437 0.073414 -0.080885 0.068634 0.013300 -0.092533 -0.018446 -0.080770 -0.078781 0.021508 -0.015698 0.062437 0.053597 -0.063038 0.001715 0.055785 -0.039040 -0.071381
|
||||
-0.026354 -0.052503 0.027507 0.058499 0.274687 -0.041930 -0.085198 -0.056231 -0.028839 -0.067498 -0.185807 -0.042463 -0.062363 0.158755 -0.060772 0.089857 0.026691 -0.051203 0.041633 0.003095 -0.026404 -0.060004 -0.095832 -0.012562 0.039131 -0.063598 -0.115453 0.095971 -0.119233 -0.039507
|
||||
-0.032215 0.002532 0.049309 0.043585 0.238954 -0.016308 -0.086309 -0.057460 -0.032669 -0.081661 -0.032760 0.040641 -0.022658 0.109967 -0.038435 0.039076 -0.012354 -0.053933 0.017739 -0.057454 -0.025240 -0.021745 -0.023133 0.019193 0.022464 0.042062 -0.127528 0.094523 -0.085705 -0.051262
|
||||
-0.063672 -0.024082 0.092149 0.025700 0.156711 0.034002 -0.079642 -0.055276 -0.054671 -0.058007 0.056388 0.082306 0.001485 0.056641 -0.000306 -0.000920 0.005379 -0.008464 -0.005483 -0.061036 -0.016189 0.033836 0.004116 -0.020099 -0.013899 0.077379 -0.054193 0.049810 -0.006214 -0.017868
|
||||
-0.114873 -0.017946 0.121477 0.004640 0.079671 0.097418 -0.060013 -0.010595 -0.072383 -0.008388 0.082942 0.061663 0.005892 -0.025134 0.011861 -0.032586 0.032611 0.084530 0.002100 -0.020639 0.007010 0.019946 0.030113 -0.017217 -0.006334 0.046930 0.013312 -0.003984 0.002495 0.005458
|
||||
-0.161713 0.036804 0.132223 0.003146 0.018287 0.125055 -0.045028 0.018140 -0.055193 0.027281 0.098250 0.043157 0.032736 -0.091279 -0.004958 -0.049990 0.053276 0.121951 0.015398 0.019613 0.017238 -0.016266 -0.002556 0.047021 -0.017725 -0.023010 0.050302 -0.015525 0.019842 0.060720
|
||||
-0.197473 0.110111 0.120398 0.007391 -0.031582 0.093153 -0.021898 0.020230 -0.017851 0.045035 0.086721 0.029438 0.052576 -0.105257 -0.033246 -0.042674 0.036293 0.092862 0.026553 0.028219 0.004797 -0.030881 -0.074813 0.053443 -0.026511 -0.065759 0.044877 -0.003945 0.038037 0.099186
|
||||
-0.199969 0.165788 0.110082 -0.004789 -0.051821 0.031355 -0.007567 -0.035291 0.029553 0.025722 0.050284 0.008884 0.056974 -0.071686 -0.038090 0.004839 -0.016414 0.031815 0.005949 0.019357 0.003543 -0.008347 -0.098634 0.021489 -0.011986 -0.035709 0.023491 -0.033810 0.022480 0.073449
|
||||
-0.167869 0.180658 0.099559 -0.034799 -0.049680 -0.036429 0.004647 -0.127286 0.050381 0.022464 0.018921 -0.009564 0.025678 0.016667 -0.024654 0.057691 -0.069632 0.001448 -0.023741 0.017199 -0.026057 0.057052 -0.048694 -0.062688 -0.008816 0.030878 0.007259 -0.023263 -0.008482 -0.001189
|
||||
-0.130954 0.183742 0.095354 -0.062048 -0.044137 -0.095424 0.003341 -0.208966 0.060799 0.055058 0.007220 -0.032092 -0.030257 0.116989 -0.006309 0.109845 -0.108615 0.018162 -0.054159 0.039139 -0.081703 0.104814 0.022952 -0.139144 0.018832 0.076335 -0.004454 -0.006528 -0.090810 -0.095068
|
||||
-0.172160 -0.109927 0.059740 -0.001614 0.039818 -0.014460 -0.039610 0.034901 0.080713 -0.036247 0.103727 -0.033208 -0.169434 0.247337 -0.041876 -0.125255 -0.114848 0.039895 0.004164 -0.084322 -0.159677 0.048388 0.051293 -0.023947 0.112857 0.153517 0.041046 -0.006995 0.080287 0.014792
|
||||
-0.135198 -0.062752 0.013771 0.014787 0.032795 -0.006058 -0.021059 0.029373 0.026822 -0.048472 0.080595 -0.021820 -0.044075 0.165348 -0.029002 -0.074665 -0.075244 0.011589 0.010367 -0.074738 -0.103821 0.028241 -0.005728 -0.061086 0.100811 0.080125 0.029016 -0.013949 0.037529 0.008324
|
||||
-0.095515 -0.016528 -0.024351 0.030765 0.024913 0.012079 -0.012422 0.026518 -0.003997 -0.054723 0.055503 -0.011654 0.045417 0.101529 -0.025318 -0.037537 -0.025453 -0.009239 0.030278 -0.061894 -0.064285 0.010310 -0.052396 -0.081756 0.076025 0.011483 0.010863 -0.010729 0.019641 0.007429
|
||||
-0.060423 0.017210 -0.054034 0.045265 0.014023 0.035755 -0.004354 0.018576 -0.017254 -0.060590 0.020824 -0.008359 0.095678 0.037622 -0.021704 -0.006522 0.035970 -0.033537 0.052688 -0.036741 -0.035840 -0.027931 -0.094169 -0.079403 0.042997 -0.070151 -0.029216 -0.002096 0.009389 0.031239
|
||||
-0.028657 0.029431 -0.068325 0.054601 0.012282 0.062780 -0.002133 0.013389 -0.025454 -0.056386 -0.011356 0.002951 0.106448 -0.028775 -0.003876 0.024019 0.108897 -0.074610 0.073446 -0.011707 -0.021342 -0.103981 -0.127299 -0.041136 0.015961 -0.150684 -0.060020 -0.007779 0.011458 0.104385
|
||||
0.063578 -0.013445 -0.036971 0.039821 0.001675 0.036406 0.008455 0.014806 0.012789 -0.060290 0.005833 -0.017255 0.148953 -0.054571 -0.045515 -0.054626 0.052297 -0.089326 0.045593 -0.031234 -0.065034 -0.020746 -0.060929 -0.009637 0.100149 0.091730 0.043765 -0.056615 0.079685 0.025615
|
||||
0.071794 -0.035989 -0.017350 0.036990 -0.003460 0.016260 0.031031 0.060910 0.012329 -0.084441 0.041786 0.005125 0.109961 -0.055672 -0.038732 -0.010692 0.032560 -0.064027 0.032960 -0.029972 -0.036901 0.013067 -0.042376 0.010142 0.063392 0.038346 0.015748 -0.029226 0.050392 0.010939
|
||||
0.063759 -0.072482 0.011329 0.033642 -0.003268 -0.003311 0.050037 0.092532 0.026586 -0.086094 0.079641 0.022486 0.031071 -0.042563 -0.036040 0.024477 0.005832 -0.040663 0.028644 -0.032777 0.009618 0.022358 -0.035611 0.038247 0.053143 -0.007266 0.004593 -0.017910 0.040641 -0.012859
|
||||
0.051630 -0.117181 0.047808 0.024724 -0.004592 -0.020789 0.064649 0.102103 0.068610 -0.083927 0.110570 0.027264 -0.082106 -0.026394 -0.033806 0.052063 -0.018428 -0.022571 0.027617 -0.038539 0.051436 0.017891 -0.037344 0.066231 0.062368 -0.049007 -0.004398 0.002072 0.022906 -0.043357
|
||||
0.037553 -0.159331 0.094129 0.007582 -0.004139 -0.035834 0.050274 0.093946 0.144177 -0.068344 0.140700 0.034470 -0.232073 -0.011532 -0.037598 0.072213 -0.029598 0.003506 0.025876 -0.030173 0.107089 -0.017808 -0.016450 0.105897 0.059851 -0.155405 -0.048537 0.057366 0.015719 -0.052152
|
||||
0.087870 0.132848 -0.106554 0.114050 0.038210 0.044132 0.032271 -0.094621 0.019907 -0.012629 -0.009526 -0.097707 -0.099677 -0.074690 -0.098412 -0.073850 0.139562 -0.146851 0.008558 -0.023334 -0.026050 -0.004419 -0.012296 0.065288 0.086510 -0.052834 0.013366 0.022267 -0.018565 0.068817
|
||||
0.086836 0.145955 -0.069460 0.111975 0.043983 0.032422 0.100857 -0.081404 0.025667 0.022035 0.046275 -0.079727 -0.054927 -0.019261 -0.090570 -0.059888 0.109093 -0.094090 0.031810 0.034953 -0.041590 0.057182 0.024488 0.110023 -0.000851 -0.006638 0.052093 0.039009 -0.033395 0.022106
|
||||
0.084080 0.154417 -0.034041 0.105690 0.045131 0.023930 0.166083 -0.067288 0.030660 0.053197 0.097038 -0.065522 -0.001021 0.037451 -0.071101 -0.036729 0.070828 -0.034174 0.046568 0.083892 -0.054168 0.115358 0.046615 0.142656 -0.079375 0.029401 0.098207 0.048457 -0.065300 -0.025684
|
||||
0.081261 0.160882 -0.003245 0.092231 0.029959 0.015570 0.223819 -0.046800 0.037761 0.072737 0.155678 -0.060748 0.055695 0.089959 -0.052061 -0.015172 0.043441 0.057430 0.065570 0.130258 -0.061637 0.173404 0.050991 0.193601 -0.155228 0.070612 0.129626 0.047902 -0.127429 -0.069416
|
||||
0.036615 0.112465 -0.011870 0.050560 -0.023586 0.056578 0.075251 -0.036631 -0.016501 0.076340 0.043630 -0.054487 0.133027 0.086040 0.092223 0.053764 -0.056705 0.064417 -0.056585 -0.137660 0.031413 -0.033211 0.145916 -0.050020 0.094662 0.028183 -0.158190 0.014984 -0.105505 0.057209
|
||||
0.034375 0.077035 -0.000712 0.093236 0.006978 0.053717 0.099504 -0.040084 -0.015700 0.079458 0.061602 -0.060225 0.054172 0.136687 0.090498 0.097151 -0.051423 0.032246 -0.057680 -0.057839 0.069192 -0.083088 0.077574 -0.067482 0.030677 -0.009552 -0.072160 0.006569 0.043961 0.017029
|
||||
0.040684 0.068853 0.009671 0.113473 0.027127 0.051794 0.115783 -0.030462 -0.025635 0.092955 0.099086 -0.070271 0.013242 0.179470 0.140747 0.141755 -0.029324 0.017410 -0.028728 -0.024224 0.109705 -0.144193 0.024627 -0.046970 -0.026776 -0.014468 -0.040679 -0.023585 0.102728 -0.026368
|
||||
0.046633 0.057103 0.003683 0.104693 0.012417 0.057466 0.110430 -0.019549 -0.020361 0.050433 0.074178 -0.043027 0.016022 0.142928 0.125023 0.156119 -0.034944 0.003438 -0.062517 -0.058566 0.054831 -0.067995 0.003152 -0.089270 -0.000192 -0.033824 -0.040037 -0.070553 0.048853 -0.033546
|
||||
0.061039 0.064860 -0.005290 0.078749 -0.007805 0.061597 0.110715 0.013244 -0.030184 0.003642 0.055152 -0.030634 0.049493 0.116908 0.140058 0.174449 -0.053516 0.000190 -0.112349 -0.132149 0.000793 -0.040724 -0.061891 -0.068736 0.005926 -0.105995 -0.060387 -0.170157 -0.047231 -0.040790
|
||||
-0.175799 -0.045261 0.024858 0.038370 0.010499 -0.010330 0.018772 -0.031099 -0.053961 0.038583 -0.102321 0.087246 -0.049020 -0.044092 0.069199 0.011624 -0.010185 -0.018390 -0.149062 0.027037 0.007544 -0.016282 0.047502 -0.012530 -0.120693 -0.109381 0.017728 0.021220 -0.018403 0.033139
|
||||
-0.150738 -0.038016 -0.005976 0.025580 -0.000282 0.015390 -0.026600 -0.045379 -0.036542 0.013962 -0.087331 0.069087 -0.008413 -0.078009 -0.005205 -0.033280 0.049669 0.006012 -0.091357 0.053438 0.023519 0.013651 0.033399 -0.008797 -0.063309 0.003493 -0.051421 0.038373 -0.008723 -0.024371
|
||||
-0.136547 -0.040596 -0.023608 0.032877 -0.009103 0.034969 -0.039678 -0.056986 -0.024972 -0.003964 -0.084974 0.047129 -0.022096 -0.085478 -0.024092 -0.032102 0.079757 -0.016028 -0.065426 0.071981 0.047015 0.004380 0.016638 -0.024282 -0.037161 0.028609 -0.064721 0.047046 -0.011783 -0.024815
|
||||
-0.090047 -0.016193 -0.016122 0.026928 -0.022206 0.045202 -0.035464 -0.047036 0.002742 0.014525 -0.052179 0.016313 -0.012168 -0.076621 0.002176 -0.051683 0.077661 -0.049370 -0.001598 0.072597 0.042840 -0.010636 0.001025 -0.024299 -0.036344 -0.050376 -0.047760 0.035163 -0.038610 -0.003729
|
||||
-0.131264 -0.034580 -0.013268 0.034774 -0.006934 0.049566 -0.031704 -0.047514 -0.010250 0.009128 -0.078666 0.042389 -0.039003 -0.101736 0.018542 -0.024093 0.062698 -0.027135 -0.071540 0.073475 0.058385 -0.017673 0.017361 -0.002414 -0.023931 -0.011269 -0.073172 0.043355 -0.015340 -0.000193
|
||||
-0.154356 -0.033249 0.000233 0.028814 0.002312 0.031366 -0.019125 -0.041617 -0.027003 0.020765 -0.078860 0.065972 -0.018866 -0.090342 0.032012 -0.028281 0.033862 -0.002061 -0.102580 0.057186 0.038928 -0.004872 0.035828 0.019269 -0.055821 -0.033986 -0.059101 0.037779 -0.010706 -0.014104
|
||||
0.033417 -0.072532 0.012568 0.008946 -0.019681 0.027342 -0.018422 -0.021703 -0.005752 0.030767 -0.102117 -0.038138 0.007722 -0.038441 -0.028165 -0.047029 0.045147 -0.041711 -0.000110 0.026793 -0.045698 0.000772 0.024380 0.000760 0.004080 0.014080 0.027246 -0.013313 -0.023959 0.002622
|
||||
0.039215 -0.137846 0.026101 0.007934 -0.015806 0.010599 -0.027141 -0.026963 -0.019964 0.048780 -0.132766 -0.032062 0.004032 -0.028996 -0.063142 -0.040783 0.045465 0.014486 -0.001545 -0.018229 -0.030509 0.038953 0.014241 -0.004312 -0.039158 0.053120 0.021326 -0.045180 0.027409 0.009402
|
||||
0.043521 -0.145592 0.040519 0.010506 -0.013667 -0.006286 -0.010463 -0.020166 -0.031415 0.042002 -0.135580 0.015098 0.024166 -0.036230 -0.025575 -0.040553 0.024242 0.005260 0.004310 -0.036449 -0.001093 0.037421 0.040208 -0.003964 -0.056208 0.059177 -0.009590 -0.033246 0.021174 -0.013008
|
||||
0.046801 -0.169637 0.077587 0.023551 0.002886 -0.035031 0.018323 -0.056422 -0.055888 0.036378 -0.131809 0.063178 0.042467 0.002594 0.074243 -0.025358 -0.016624 -0.042451 0.017027 -0.058671 0.046846 0.011740 0.094474 -0.078083 -0.084720 0.033242 -0.013165 -0.023825 -0.006214 -0.047684
|
||||
0.047127 -0.148675 0.043016 0.015841 -0.008682 0.009800 -0.012490 -0.031526 -0.026918 0.043809 -0.147540 0.010436 0.022248 -0.041771 0.015715 -0.044134 0.019909 0.003267 0.012343 -0.043093 0.007270 0.028899 0.042830 0.005583 -0.024609 0.043346 -0.020205 -0.020537 0.014459 -0.010947
|
||||
0.037862 -0.132432 0.028533 0.015229 -0.009476 0.027310 -0.028606 -0.036439 -0.009620 0.058203 -0.140411 -0.033794 0.004687 -0.041138 -0.009726 -0.051973 0.027422 -0.006463 0.000827 -0.017258 -0.011153 0.018941 0.007685 0.000585 -0.000785 0.036124 0.010801 -0.031023 0.025637 0.006657
|
||||
-0.072053 -0.074598 -0.100711 -0.160392 -0.056590 -0.046253 0.022243 -0.033939 -0.070682 0.062764 0.074060 -0.097748 0.012162 0.038627 0.038864 0.066326 -0.035816 -0.139004 0.083591 0.117267 0.134277 0.020804 0.001839 0.085517 0.010905 0.060183 -0.029659 0.033860 0.004833 0.047905
|
||||
0.008856 0.022312 -0.093092 -0.096187 -0.110312 -0.030930 -0.023096 -0.007133 -0.029338 0.005534 0.021892 -0.041877 0.024922 -0.025176 0.009405 0.027628 -0.064749 0.029258 0.045688 0.051081 0.010396 -0.039200 -0.007260 -0.072501 0.003076 0.004612 0.027440 -0.007110 -0.028578 -0.055429
|
||||
0.042419 0.075052 -0.081836 -0.028702 -0.141576 -0.019808 -0.044103 0.022094 0.023906 -0.041791 0.007334 0.020262 0.009451 -0.052770 -0.013873 -0.032374 -0.045137 0.114123 0.003529 0.025807 -0.060978 -0.045002 -0.069777 -0.072668 -0.059290 -0.013371 0.048115 0.047343 -0.008194 -0.095606
|
||||
0.056245 0.073217 -0.085604 -0.014481 -0.139289 -0.012952 -0.065166 0.021197 0.040232 -0.054156 -0.003544 0.024890 -0.013008 -0.038209 0.002685 -0.040139 -0.023651 0.089468 0.003185 0.028247 -0.061535 -0.064831 -0.076752 -0.083293 -0.081248 0.035841 -0.022897 0.061927 0.006909 -0.072040
|
||||
0.056430 0.054020 -0.075907 -0.028402 -0.150533 -0.007712 -0.061734 0.021080 0.030544 -0.035952 0.013622 0.019362 0.007588 -0.014823 0.006822 -0.068151 -0.027590 0.106978 0.012945 0.046588 -0.080258 -0.070291 -0.051863 -0.083578 -0.101646 0.031158 -0.023454 0.089560 -0.024262 -0.077726
|
||||
0.024479 -0.009246 -0.089056 -0.082610 -0.118332 -0.003018 -0.047311 0.022264 -0.024587 -0.005407 0.027436 -0.026817 -0.000381 0.000645 0.030356 -0.021404 -0.042803 -0.008602 0.024542 0.091667 -0.051169 -0.082852 -0.054013 -0.049463 -0.075523 0.039438 -0.024571 -0.002215 -0.035084 -0.052036
|
||||
-0.012027 -0.136883 -0.081754 -0.149149 -0.055287 -0.009700 0.008094 -0.006322 -0.083587 0.062632 0.030406 -0.119596 -0.044029 0.034376 0.058444 0.083058 0.008314 -0.131333 0.060136 0.135252 0.037775 0.014841 -0.105155 0.067016 -0.042349 0.057074 0.020755 -0.044335 0.016350 0.028098
|
||||
0.012816 -0.053653 -0.100099 -0.097988 0.007326 -0.063234 -0.044976 0.019100 0.000113 0.020415 0.014232 -0.015097 -0.037545 0.018025 0.050639 -0.026776 0.001709 -0.055574 0.006984 0.107057 -0.015655 -0.009003 0.051013 0.002343 0.006298 -0.027962 0.015854 -0.091353 -0.016346 0.039465
|
||||
0.035534 0.010285 -0.097093 -0.025993 0.043429 -0.098898 -0.077835 0.037713 0.075033 -0.019078 -0.028117 0.072858 -0.038120 -0.010628 0.052424 -0.079396 0.009461 0.000612 -0.023804 0.043805 -0.006746 -0.000033 0.127073 -0.008097 0.050749 -0.034756 0.026520 -0.083538 -0.050395 0.020636
|
||||
0.030421 0.040185 -0.107600 -0.020036 0.045443 -0.101940 -0.073577 0.039398 0.089641 -0.036760 -0.037780 0.091039 -0.036830 -0.033786 0.025635 -0.067609 -0.020951 -0.002102 -0.021006 0.006076 0.043247 0.015986 0.090187 0.005710 0.062335 -0.028713 0.040719 -0.131139 -0.025086 0.002108
|
||||
0.011036 0.047297 -0.102586 -0.035860 0.046121 -0.110718 -0.049199 0.029296 0.077701 -0.015091 -0.022442 0.073219 -0.003970 -0.046326 0.008115 -0.052553 -0.037889 0.005420 -0.025423 -0.027524 0.058565 0.027155 0.082429 0.016126 0.080709 -0.088126 0.109562 -0.159245 -0.006448 -0.016464
|
||||
-0.023863 -0.001767 -0.112029 -0.110537 0.012159 -0.090593 -0.009916 0.004722 0.005667 0.023478 0.015308 -0.012738 0.005787 -0.017643 0.017190 0.018514 -0.050562 -0.056536 0.012880 0.004994 0.085020 0.028885 0.071570 0.002042 0.074510 -0.049359 0.066226 -0.121898 -0.005946 0.006471
|
||||
-0.050886 -0.055706 -0.121851 -0.167249 -0.044477 -0.046575 0.028548 -0.042874 -0.061687 0.059400 0.060959 -0.100667 0.005081 0.035377 0.021717 0.060243 -0.030080 -0.150824 0.080442 0.090665 0.113206 0.029119 0.010088 0.105332 -0.006041 0.123141 -0.083657 0.047525 0.028597 0.091883
|
||||
0.030708 0.078089 -0.080455 -0.012260 -0.109552 -0.041103 -0.041066 0.009557 0.026507 -0.046242 -0.026526 0.046953 0.028301 -0.060457 0.018475 -0.053304 -0.043608 0.094731 0.031432 -0.117133 -0.076221 -0.030731 0.016484 0.046091 -0.039803 -0.002365 0.051830 0.126565 0.100037 -0.043647
|
||||
0.055139 0.076245 -0.081522 0.004314 -0.107992 -0.039650 -0.063847 0.014757 0.044175 -0.065309 -0.027112 0.053595 -0.026360 -0.046824 0.012100 -0.022947 -0.031940 0.048766 0.001973 -0.055875 -0.071290 -0.061669 -0.018830 -0.025187 -0.043714 0.001550 0.010165 0.155243 0.019864 -0.063997
|
||||
0.060624 0.046756 -0.072069 -0.012080 -0.112141 -0.035260 -0.050129 0.028093 0.012172 -0.057691 -0.002134 0.044817 -0.014066 -0.049732 0.013478 0.000195 -0.048403 0.116300 -0.003782 -0.070723 -0.092908 -0.053813 0.005494 -0.043911 -0.042773 -0.024812 0.028047 0.200589 0.002444 0.028023
|
||||
-0.002693 -0.109254 -0.103698 -0.155319 -0.040843 -0.008397 0.025152 -0.004684 -0.075416 0.044165 0.036425 -0.106068 -0.063753 0.008950 0.032720 0.107046 0.006529 -0.150703 0.049184 0.114889 0.065819 0.022378 -0.105813 0.077860 -0.041574 0.066244 0.002608 -0.063551 0.013985 0.008026
|
||||
0.056392 0.032564 -0.123103 -0.042145 0.014634 -0.085346 -0.045427 0.044037 0.054277 -0.044440 -0.015580 0.091353 -0.020715 -0.046674 0.024152 -0.008198 -0.036096 0.105219 -0.027204 -0.058790 0.022864 0.002164 0.063033 -0.012096 0.021425 -0.015333 0.011114 -0.013654 0.020932 0.125969
|
||||
0.050734 0.066831 -0.133624 -0.025020 0.025214 -0.086764 -0.052195 0.036576 0.090356 -0.053293 -0.040101 0.097309 -0.028327 -0.042717 0.020156 -0.038668 -0.031980 0.037324 -0.028638 -0.046635 0.037633 0.003905 0.057016 0.012124 0.023034 0.010380 0.012591 -0.047595 0.045870 0.056652
|
||||
0.024006 0.067660 -0.131503 -0.044887 0.014372 -0.089393 -0.029303 0.029039 0.070116 -0.027935 -0.042161 0.091030 0.024491 -0.049115 0.023052 -0.069348 -0.035778 0.081727 -0.007838 -0.092164 0.033586 0.027022 0.084316 0.076691 0.038307 0.018769 0.032070 -0.086649 0.139818 0.057912
|
||||
# The variances of the components (eigenvalues) of identity or combined identity and expression model
|
||||
1
|
||||
30
|
||||
6
|
||||
1260.970668 622.171792 514.701969 368.231621 178.288803 141.373501 123.600203 108.077802 95.536656 70.169455 60.351993 56.235863 49.519216 44.933492 42.272445 39.564849 34.747298 28.830661 26.147964 25.285930 21.833110 18.995127 18.134667 16.329230 13.965430 12.933174 11.705294 11.301406 9.624030 8.423463
|
||||
Binary file not shown.
@@ -0,0 +1,29 @@
|
||||
% Script for internal menpo data split, convert the original training data
|
||||
% to 2/3rds training and 1/3rd validation
|
||||
|
||||
menpo_root = 'D:\Datasets\menpo/';
|
||||
|
||||
pts_files = dir([menpo_root, '/*.jpg']);
|
||||
jpg_files = dir([menpo_root, '/*.pts']);
|
||||
|
||||
% Do the actual copying to respective folders
|
||||
out_train = [menpo_root, '/train/'];
|
||||
out_valid = [menpo_root, '/valid/'];
|
||||
mkdir(out_train);
|
||||
mkdir(out_valid);
|
||||
|
||||
load('train_valid_split');
|
||||
|
||||
for i=1:numel(train_imgs)
|
||||
|
||||
copyfile([menpo_root, train_pts(i).name], [out_train, train_pts(i).name]);
|
||||
copyfile([menpo_root, train_imgs(i).name], [out_train, train_imgs(i).name]);
|
||||
|
||||
end
|
||||
|
||||
for i=1:numel(valid_imgs)
|
||||
|
||||
copyfile([menpo_root, valid_pts(i).name], [out_valid, valid_pts(i).name]);
|
||||
copyfile([menpo_root, valid_imgs(i).name], [out_valid, valid_imgs(i).name]);
|
||||
|
||||
end
|
||||
Binary file not shown.
Binary file not shown.
@@ -0,0 +1,41 @@
|
||||
function writePDM( V, E, M, outputFile, Vmorph, Emorph )
|
||||
%WRITEPDM Summary of this function goes here
|
||||
% Detailed explanation goes here
|
||||
|
||||
fId = fopen(outputFile,'w');
|
||||
|
||||
% number of elements
|
||||
|
||||
% Comment
|
||||
fprintf(fId, '# The mean values of the components (in mm)\n');
|
||||
writeMatrix(fId, M, 6);
|
||||
|
||||
fprintf(fId, '# The principal components (eigenvectors) of identity or combined identity and expression model\n');
|
||||
writeMatrix(fId, V, 6);
|
||||
|
||||
fprintf(fId, '# The variances of the components (eigenvalues) of identity or combined identity and expression model\n');
|
||||
writeMatrix(fId, E', 6);
|
||||
|
||||
if(nargin > 4)
|
||||
fprintf(fId, '# The principal components (eigenvectors) of expression\n');
|
||||
writeMatrix(fId, Vmorph, 6);
|
||||
|
||||
fprintf(fId, '# The variances of the components (eigenvalues) of expression\n');
|
||||
writeMatrix(fId, Emorph', 6);
|
||||
end
|
||||
|
||||
fclose(fId);
|
||||
end
|
||||
|
||||
% for easier readibility write them row by row
|
||||
function writeMatrix(fileID, M, type)
|
||||
|
||||
fprintf(fileID, '%d\n', size(M,1));
|
||||
fprintf(fileID, '%d\n', size(M,2));
|
||||
fprintf(fileID, '%d\n', type);
|
||||
|
||||
for i=1:size(M,1)
|
||||
fprintf(fileID, '%f ', M(i,:));
|
||||
fprintf(fileID, '\n');
|
||||
end
|
||||
end
|
||||
Reference in New Issue
Block a user