notebook.community

Edit and run 





In [1]:



    


%matplotlib inline
%pylab inline



    


















    






Populating the interactive namespace from numpy and matplotlib

















In [2]:



    


repeat = 1

Load test data





In [3]:



    


import menpo.io as mio
from menpo.landmark import labeller, lfpw_face
from menpofast.utils import convert_from_menpo

path = '/data/'
group = 'lfpw_face'

test_images = []
for i in mio.import_images(path + 'PhD/DataBases/faces/cofw/testset/', verbose=True, 
                           max_images=None):
    
    # convert the image from menpo Image to menpofast Image (channels at front)
    i = convert_from_menpo(i)
    
    labeller(i, 'PTS', eval(group))
    i.crop_to_landmarks_proportion_inplace(1, group='PTS')
    
    if i.n_channels == 3:
        i = i.as_greyscale(mode='average')
    test_images.append(i)



    


















    






- Loading 507 assets: [====================] 100%
















In [4]:



    


from menpo.visualize import visualize_images

visualize_images(test_images)

AAMs

Load





In [5]:



    


from alabortcvpr2015.utils import pickle_load

aam = pickle_load(path + 'PhD/Models/aam_cofw_fast_dsift')

AIC





In [6]:



    


sampling_mask = np.require(np.zeros(aam.parts_shape), dtype=np.bool)
sampling_mask[3::7, 3::7] = True

imshow(sampling_mask)



    


















    
Out[6]:








<matplotlib.image.AxesImage at 0x7fcdc4f49490>










    
























In [13]:



    


from alabortcvpr2015.aam import PartsAAMFitter, AIC
from alabortcvpr2015.utils import pickle_dump
from alabortcvpr2015.result import SerializableResult

fitter = PartsAAMFitter(aam, algorithm_cls=AIC, n_shape=[3, 12],
                        n_appearance=[25, 50], sampling_mask=sampling_mask)

fitter_results = []

for seed in xrange(repeat):

    np.random.seed(seed=seed)

    for j, i in enumerate(test_images):

        gt_s = i.landmarks[group].lms
        s = fitter.perturb_shape(gt_s, noise_std=0.05)

        fr = fitter.fit(i, s, gt_shape=gt_s, max_iters=20, prior=False)

        fitter_results.append(fr)
        fr.downscale = 0.5

        print 'Image: ', j
        print fr

results = [SerializableResult('none', fr.shapes(), fr.n_iters, 'AIC', fr.gt_shape) 
           for fr in fitter_results]

pickle_dump(results, path + 'PhD/Results/aam_aic_cofw_fast_dsift')



    


















    






Image:  0
Initial error: 0.1455
Final error: 0.0618
Image:  1
Initial error: 0.1943
Final error: 0.0609
Image:  2
Initial error: 0.1043
Final error: 0.0289
Image:  3
Initial error: 0.0740
Final error: 0.0420
Image:  4
Initial error: 0.0568
Final error: 0.0369
Image:  5
Initial error: 0.0896
Final error: 0.0220
Image:  6
Initial error: 0.0515
Final error: 0.0276
Image:  7
Initial error: 0.0844
Final error: 0.0365
Image:  8
Initial error: 0.0743
Final error: 0.0363
Image:  9
Initial error: 0.0349
Final error: 0.0291
Image:  10
Initial error: 0.1329
Final error: 0.0373
Image:  11
Initial error: 0.0988
Final error: 0.0236
Image:  12
Initial error: 0.1064
Final error: 0.0462
Image:  13
Initial error: 0.0610
Final error: 0.0782
Image:  14
Initial error: 0.0597
Final error: 0.0353
Image:  15
Initial error: 0.1343
Final error: 0.0391
Image:  16
Initial error: 0.0953
Final error: 0.0242
Image:  17
Initial error: 0.0350
Final error: 0.0175
Image:  18
Initial error: 0.0438
Final error: 0.0428
Image:  19
Initial error: 0.0525
Final error: 0.0242
Image:  20
Initial error: 0.1728
Final error: 0.0867
Image:  21
Initial error: 0.0794
Final error: 0.0204
Image:  22
Initial error: 0.0570
Final error: 0.0267
Image:  23
Initial error: 0.0569
Final error: 0.0277
Image:  24
Initial error: 0.1400
Final error: 0.0219
Image:  25
Initial error: 0.0925
Final error: 0.0447
Image:  26
Initial error: 0.0780
Final error: 0.0355
Image:  27
Initial error: 0.1338
Final error: 0.0447
Image:  28
Initial error: 0.0598
Final error: 0.0245
Image:  29
Initial error: 0.0710
Final error: 0.0199
Image:  30
Initial error: 0.0756
Final error: 0.0393
Image:  31
Initial error: 0.0886
Final error: 0.0396
Image:  32
Initial error: 0.1199
Final error: 0.0283
Image:  33
Initial error: 0.0646
Final error: 0.0241
Image:  34
Initial error: 0.0666
Final error: 0.0310
Image:  35
Initial error: 0.0633
Final error: 0.0256
Image:  36
Initial error: 0.1225
Final error: 0.0393
Image:  37
Initial error: 0.1070
Final error: 0.0278
Image:  38
Initial error: 0.2081
Final error: 0.0444
Image:  39
Initial error: 0.1125
Final error: 0.0415
Image:  40
Initial error: 0.1040
Final error: 0.0266
Image:  41
Initial error: 0.1221
Final error: 0.0295
Image:  42
Initial error: 0.0702
Final error: 0.0231
Image:  43
Initial error: 0.0794
Final error: 0.0622
Image:  44
Initial error: 0.0574
Final error: 0.0253
Image:  45
Initial error: 0.1360
Final error: 0.0469
Image:  46
Initial error: 0.0614
Final error: 0.0321
Image:  47
Initial error: 0.1019
Final error: 0.0629
Image:  48
Initial error: 0.0547
Final error: 0.0497
Image:  49
Initial error: 0.1382
Final error: 0.0306
Image:  50
Initial error: 0.0935
Final error: 0.0386
Image:  51
Initial error: 0.0507
Final error: 0.0277
Image:  52
Initial error: 0.0813
Final error: 0.0565
Image:  53
Initial error: 0.0541
Final error: 0.0233
Image:  54
Initial error: 0.1199
Final error: 0.0375
Image:  55
Initial error: 0.0797
Final error: 0.0235
Image:  56
Initial error: 0.0550
Final error: 0.0581
Image:  57
Initial error: 0.2140
Final error: 0.2351
Image:  58
Initial error: 0.0827
Final error: 0.0250
Image:  59
Initial error: 0.2586
Final error: 0.2386
Image:  60
Initial error: 0.0609
Final error: 0.0464
Image:  61
Initial error: 0.0739
Final error: 0.0275
Image:  62
Initial error: 0.0645
Final error: 0.0374
Image:  63
Initial error: 0.1605
Final error: 0.0428
Image:  64
Initial error: 0.1035
Final error: 0.0225
Image:  65
Initial error: 0.1046
Final error: 0.0563
Image:  66
Initial error: 0.0749
Final error: 0.0419
Image:  67
Initial error: 0.1410
Final error: 0.0460
Image:  68
Initial error: 0.1158
Final error: 0.0504
Image:  69
Initial error: 0.1469
Final error: 0.0200
Image:  70
Initial error: 0.0782
Final error: 0.0224
Image:  71
Initial error: 0.0872
Final error: 0.0400
Image:  72
Initial error: 0.0511
Final error: 0.0363
Image:  73
Initial error: 0.0904
Final error: 0.0413
Image:  74
Initial error: 0.0529
Final error: 0.0264
Image:  75
Initial error: 0.0771
Final error: 0.0399
Image:  76
Initial error: 0.1009
Final error: 0.0219
Image:  77
Initial error: 0.0674
Final error: 0.0292
Image:  78
Initial error: 0.0708
Final error: 0.0361
Image:  79
Initial error: 0.0310
Final error: 0.0199
Image:  80
Initial error: 0.0736
Final error: 0.0432
Image:  81
Initial error: 0.1060
Final error: 0.0302
Image:  82
Initial error: 0.0856
Final error: 0.0456
Image:  83
Initial error: 0.1551
Final error: 0.0789
Image:  84
Initial error: 0.0984
Final error: 0.0311
Image:  85
Initial error: 0.0657
Final error: 0.0192
Image:  86
Initial error: 0.0528
Final error: 0.0240
Image:  87
Initial error: 0.0888
Final error: 0.0261
Image:  88
Initial error: 0.0645
Final error: 0.0353
Image:  89
Initial error: 0.0694
Final error: 0.0304
Image:  90
Initial error: 0.1102
Final error: 0.0734
Image:  91
Initial error: 0.0489
Final error: 0.0311
Image:  92
Initial error: 0.0618
Final error: 0.0288
Image:  93
Initial error: 0.0567
Final error: 0.0330
Image:  94
Initial error: 0.0565
Final error: 0.0237
Image:  95
Initial error: 0.1755
Final error: 0.0309
Image:  96
Initial error: 0.0939
Final error: 0.0678
Image:  97
Initial error: 0.0607
Final error: 0.0321
Image:  98
Initial error: 0.0642
Final error: 0.0399
Image:  99
Initial error: 0.1066
Final error: 0.0544
Image:  100
Initial error: 0.0566
Final error: 0.0542
Image:  101
Initial error: 0.0460
Final error: 0.0237
Image:  102
Initial error: 0.1414
Final error: 0.0676
Image:  103
Initial error: 0.0969
Final error: 0.0405
Image:  104
Initial error: 0.1130
Final error: 0.0208
Image:  105
Initial error: 0.1171
Final error: 0.0318
Image:  106
Initial error: 0.1626
Final error: 0.0461
Image:  107
Initial error: 0.0946
Final error: 0.0336
Image:  108
Initial error: 0.0640
Final error: 0.0274
Image:  109
Initial error: 0.0630
Final error: 0.0417
Image:  110
Initial error: 0.0563
Final error: 0.0471
Image:  111
Initial error: 0.0975
Final error: 0.0576
Image:  112
Initial error: 0.0880
Final error: 0.0429
Image:  113
Initial error: 0.0977
Final error: 0.0349
Image:  114
Initial error: 0.0317
Final error: 0.0292
Image:  115
Initial error: 0.0545
Final error: 0.0597
Image:  116
Initial error: 0.0892
Final error: 0.1000
Image:  117
Initial error: 0.1502
Final error: 0.0437
Image:  118
Initial error: 0.1312
Final error: 0.0579
Image:  119
Initial error: 0.0860
Final error: 0.0463
Image:  120
Initial error: 0.0807
Final error: 0.0328
Image:  121
Initial error: 0.1010
Final error: 0.0478
Image:  122
Initial error: 0.0853
Final error: 0.0263
Image:  123
Initial error: 0.1709
Final error: 0.0658
Image:  124
Initial error: 0.1103
Final error: 0.0426
Image:  125
Initial error: 0.1276
Final error: 0.1070
Image:  126
Initial error: 0.1330
Final error: 0.0159
Image:  127
Initial error: 0.0568
Final error: 0.0234
Image:  128
Initial error: 0.0784
Final error: 0.0388
Image:  129
Initial error: 0.0825
Final error: 0.0381
Image:  130
Initial error: 0.1277
Final error: 0.0430
Image:  131
Initial error: 0.0869
Final error: 0.0372
Image:  132
Initial error: 0.1745
Final error: 0.0415
Image:  133
Initial error: 0.0741
Final error: 0.0302
Image:  134
Initial error: 0.0380
Final error: 0.0216
Image:  135
Initial error: 0.0680
Final error: 0.0266
Image:  136
Initial error: 0.1427
Final error: 0.0398
Image:  137
Initial error: 0.0531
Final error: 0.0259
Image:  138
Initial error: 0.0824
Final error: 0.0205
Image:  139
Initial error: 0.1574
Final error: 0.0342
Image:  140
Initial error: 0.0859
Final error: 0.0474
Image:  141
Initial error: 0.1135
Final error: 0.0326
Image:  142
Initial error: 0.0967
Final error: 0.0503
Image:  143
Initial error: 0.0709
Final error: 0.0366
Image:  144
Initial error: 0.0746
Final error: 0.0628
Image:  145
Initial error: 0.0995
Final error: 0.0281
Image:  146
Initial error: 0.1562
Final error: 0.0629
Image:  147
Initial error: 0.0296
Final error: 0.0230
Image:  148
Initial error: 0.0746
Final error: 0.0303
Image:  149
Initial error: 0.0788
Final error: 0.0449
Image:  150
Initial error: 0.0865
Final error: 0.0330
Image:  151
Initial error: 0.1486
Final error: 0.0516
Image:  152
Initial error: 0.1261
Final error: 0.0384
Image:  153
Initial error: 0.0593
Final error: 0.0798
Image:  154
Initial error: 0.1148
Final error: 0.0409
Image:  155
Initial error: 0.0424
Final error: 0.0218
Image:  156
Initial error: 0.0696
Final error: 0.0285
Image:  157
Initial error: 0.0897
Final error: 0.0285
Image:  158
Initial error: 0.0709
Final error: 0.0388
Image:  159
Initial error: 0.0770
Final error: 0.0250
Image:  160
Initial error: 0.0896
Final error: 0.0331
Image:  161
Initial error: 0.0566
Final error: 0.0408
Image:  162
Initial error: 0.1261
Final error: 0.0195
Image:  163
Initial error: 0.0371
Final error: 0.0229
Image:  164
Initial error: 0.0896
Final error: 0.0502
Image:  165
Initial error: 0.1579
Final error: 0.0606
Image:  166
Initial error: 0.0461
Final error: 0.0202
Image:  167
Initial error: 0.1395
Final error: 0.0239
Image:  168
Initial error: 0.0635
Final error: 0.0347
Image:  169
Initial error: 0.0577
Final error: 0.0336
Image:  170
Initial error: 0.0677
Final error: 0.0261
Image:  171
Initial error: 0.1362
Final error: 0.0374
Image:  172
Initial error: 0.0548
Final error: 0.0261
Image:  173
Initial error: 0.0878
Final error: 0.0287
Image:  174
Initial error: 0.0631
Final error: 0.0235
Image:  175
Initial error: 0.1121
Final error: 0.0374
Image:  176
Initial error: 0.0838
Final error: 0.0402
Image:  177
Initial error: 0.0726
Final error: 0.0440
Image:  178
Initial error: 0.0756
Final error: 0.0460
Image:  179
Initial error: 0.1107
Final error: 0.0361
Image:  180
Initial error: 0.0490
Final error: 0.0407
Image:  181
Initial error: 0.0922
Final error: 0.0531
Image:  182
Initial error: 0.0570
Final error: 0.0332
Image:  183
Initial error: 0.2303
Final error: 0.1158
Image:  184
Initial error: 0.0729
Final error: 0.0204
Image:  185
Initial error: 0.0955
Final error: 0.0341
Image:  186
Initial error: 0.0428
Final error: 0.0336
Image:  187
Initial error: 0.1089
Final error: 0.0265
Image:  188
Initial error: 0.0475
Final error: 0.0201
Image:  189
Initial error: 0.0721
Final error: 0.0253
Image:  190
Initial error: 0.0402
Final error: 0.0214
Image:  191
Initial error: 0.1259
Final error: 0.0238
Image:  192
Initial error: 0.0933
Final error: 0.0346
Image:  193
Initial error: 0.1342
Final error: 0.0502
Image:  194
Initial error: 0.0999
Final error: 0.0185
Image:  195
Initial error: 0.0704
Final error: 0.0226
Image:  196
Initial error: 0.1382
Final error: 0.0681
Image:  197
Initial error: 0.0759
Final error: 0.0204
Image:  198
Initial error: 0.0925
Final error: 0.0308
Image:  199
Initial error: 0.2003
Final error: 0.0391
Image:  200
Initial error: 0.0296
Final error: 0.0347
Image:  201
Initial error: 0.0662
Final error: 0.0304
Image:  202
Initial error: 0.1082
Final error: 0.0481
Image:  203
Initial error: 0.1051
Final error: 0.0339
Image:  204
Initial error: 0.0694
Final error: 0.0217
Image:  205
Initial error: 0.0642
Final error: 0.0332
Image:  206
Initial error: 0.1591
Final error: 0.0680
Image:  207
Initial error: 0.0743
Final error: 0.0338
Image:  208
Initial error: 0.0834
Final error: 0.0731
Image:  209
Initial error: 0.0302
Final error: 0.0488
Image:  210
Initial error: 0.1182
Final error: 0.0312
Image:  211
Initial error: 0.1392
Final error: 0.0965
Image:  212
Initial error: 0.1727
Final error: 0.0576
Image:  213
Initial error: 0.0967
Final error: 0.0360
Image:  214
Initial error: 0.0693
Final error: 0.0207
Image:  215
Initial error: 0.0467
Final error: 0.0457
Image:  216
Initial error: 0.0462
Final error: 0.0238
Image:  217
Initial error: 0.0689
Final error: 0.0182
Image:  218
Initial error: 0.0753
Final error: 0.0265
Image:  219
Initial error: 0.0913
Final error: 0.0201
Image:  220
Initial error: 0.0454
Final error: 0.0221
Image:  221
Initial error: 0.0881
Final error: 0.0332
Image:  222
Initial error: 0.0522
Final error: 0.0308
Image:  223
Initial error: 0.1234
Final error: 0.0303
Image:  224
Initial error: 0.1384
Final error: 0.0444
Image:  225
Initial error: 0.0403
Final error: 0.0319
Image:  226
Initial error: 0.0928
Final error: 0.0317
Image:  227
Initial error: 0.0977
Final error: 0.0506
Image:  228
Initial error: 0.0862
Final error: 0.0228
Image:  229
Initial error: 0.0313
Final error: 0.0369
Image:  230
Initial error: 0.0766
Final error: 0.0409
Image:  231
Initial error: 0.0461
Final error: 0.0217
Image:  232
Initial error: 0.0966
Final error: 0.0316
Image:  233
Initial error: 0.0930
Final error: 0.0623
Image:  234
Initial error: 0.0836
Final error: 0.0392
Image:  235
Initial error: 0.1278
Final error: 0.0380
Image:  236
Initial error: 0.0792
Final error: 0.0309
Image:  237
Initial error: 0.0720
Final error: 0.0459
Image:  238
Initial error: 0.0644
Final error: 0.0320
Image:  239
Initial error: 0.1429
Final error: 0.0206
Image:  240
Initial error: 0.0946
Final error: 0.0237
Image:  241
Initial error: 0.0618
Final error: 0.0219
Image:  242
Initial error: 0.0744
Final error: 0.0269
Image:  243
Initial error: 0.1122
Final error: 0.0253
Image:  244
Initial error: 0.0575
Final error: 0.0613
Image:  245
Initial error: 0.0544
Final error: 0.0570
Image:  246
Initial error: 0.0538
Final error: 0.0387
Image:  247
Initial error: 0.0780
Final error: 0.0376
Image:  248
Initial error: 0.1808
Final error: 0.1044
Image:  249
Initial error: 0.1456
Final error: 0.0337
Image:  250
Initial error: 0.0451
Final error: 0.0447
Image:  251
Initial error: 0.1166
Final error: 0.0296
Image:  252
Initial error: 0.0973
Final error: 0.0265
Image:  253
Initial error: 0.0885
Final error: 0.0275
Image:  254
Initial error: 0.0466
Final error: 0.0353
Image:  255
Initial error: 0.0679
Final error: 0.0409
Image:  256
Initial error: 0.0830
Final error: 0.0476
Image:  257
Initial error: 0.0432
Final error: 0.0176
Image:  258
Initial error: 0.0673
Final error: 0.0361
Image:  259
Initial error: 0.0865
Final error: 0.0262
Image:  260
Initial error: 0.0908
Final error: 0.0240
Image:  261
Initial error: 0.0740
Final error: 0.0309
Image:  262
Initial error: 0.0900
Final error: 0.0462
Image:  263
Initial error: 0.1658
Final error: 0.0489
Image:  264
Initial error: 0.0854
Final error: 0.0252
Image:  265
Initial error: 0.0829
Final error: 0.0237
Image:  266
Initial error: 0.0311
Final error: 0.0154
Image:  267
Initial error: 0.1021
Final error: 0.0337
Image:  268
Initial error: 0.0379
Final error: 0.0265
Image:  269
Initial error: 0.0634
Final error: 0.0458
Image:  270
Initial error: 0.0793
Final error: 0.0297
Image:  271
Initial error: 0.0943
Final error: 0.0380
Image:  272
Initial error: 0.0656
Final error: 0.0188
Image:  273
Initial error: 0.1027
Final error: 0.0400
Image:  274
Initial error: 0.0564
Final error: 0.0189
Image:  275
Initial error: 0.1014
Final error: 0.0399
Image:  276
Initial error: 0.0445
Final error: 0.0279
Image:  277
Initial error: 0.0804
Final error: 0.0397
Image:  278
Initial error: 0.1770
Final error: 0.1953
Image:  279
Initial error: 0.0711
Final error: 0.0278
Image:  280
Initial error: 0.0745
Final error: 0.0240
Image:  281
Initial error: 0.0704
Final error: 0.0336
Image:  282
Initial error: 0.0894
Final error: 0.0212
Image:  283
Initial error: 0.1841
Final error: 0.0370
Image:  284
Initial error: 0.1010
Final error: 0.0206
Image:  285
Initial error: 0.1247
Final error: 0.0509
Image:  286
Initial error: 0.1321
Final error: 0.0759
Image:  287
Initial error: 0.0635
Final error: 0.0352
Image:  288
Initial error: 0.0757
Final error: 0.0387
Image:  289
Initial error: 0.1199
Final error: 0.0329
Image:  290
Initial error: 0.0550
Final error: 0.0239
Image:  291
Initial error: 0.1876
Final error: 0.1952
Image:  292
Initial error: 0.0652
Final error: 0.0377
Image:  293
Initial error: 0.1901
Final error: 0.1489
Image:  294
Initial error: 0.1188
Final error: 0.0290
Image:  295
Initial error: 0.0774
Final error: 0.0433
Image:  296
Initial error: 0.1737
Final error: 0.0449
Image:  297
Initial error: 0.0759
Final error: 0.0391
Image:  298
Initial error: 0.1079
Final error: 0.0520
Image:  299
Initial error: 0.1248
Final error: 0.0241
Image:  300
Initial error: 0.0788
Final error: 0.0336
Image:  301
Initial error: 0.0819
Final error: 0.0237
Image:  302
Initial error: 0.0746
Final error: 0.0294
Image:  303
Initial error: 0.1413
Final error: 0.0437
Image:  304
Initial error: 0.0938
Final error: 0.0314
Image:  305
Initial error: 0.0980
Final error: 0.0848
Image:  306
Initial error: 0.0504
Final error: 0.0241
Image:  307
Initial error: 0.0532
Final error: 0.0432
Image:  308
Initial error: 0.1361
Final error: 0.0544
Image:  309
Initial error: 0.0621
Final error: 0.0207
Image:  310
Initial error: 0.0880
Final error: 0.0337
Image:  311
Initial error: 0.1025
Final error: 0.0622
Image:  312
Initial error: 0.0863
Final error: 0.0394
Image:  313
Initial error: 0.1403
Final error: 0.0307
Image:  314
Initial error: 0.1173
Final error: 0.0311
Image:  315
Initial error: 0.0981
Final error: 0.0338
Image:  316
Initial error: 0.1342
Final error: 0.0315
Image:  317
Initial error: 0.1548
Final error: 0.0298
Image:  318
Initial error: 0.0686
Final error: 0.0296
Image:  319
Initial error: 0.0933
Final error: 0.0247
Image:  320
Initial error: 0.0571
Final error: 0.0269
Image:  321
Initial error: 0.0915
Final error: 0.0405
Image:  322
Initial error: 0.1291
Final error: 0.0474
Image:  323
Initial error: 0.1200
Final error: 0.0429
Image:  324
Initial error: 0.1173
Final error: 0.0510
Image:  325
Initial error: 0.0479
Final error: 0.0216
Image:  326
Initial error: 0.0901
Final error: 0.0482
Image:  327
Initial error: 0.0582
Final error: 0.0288
Image:  328
Initial error: 0.0629
Final error: 0.0242
Image:  329
Initial error: 0.1068
Final error: 0.0538
Image:  330
Initial error: 0.0460
Final error: 0.0254
Image:  331
Initial error: 0.0775
Final error: 0.0521
Image:  332
Initial error: 0.1022
Final error: 0.0432
Image:  333
Initial error: 0.1114
Final error: 0.0354
Image:  334
Initial error: 0.0993
Final error: 0.0754
Image:  335
Initial error: 0.0867
Final error: 0.0301
Image:  336
Initial error: 0.1043
Final error: 0.0512
Image:  337
Initial error: 0.1331
Final error: 0.0563
Image:  338
Initial error: 0.1272
Final error: 0.1068
Image:  339
Initial error: 0.1505
Final error: 0.0405
Image:  340
Initial error: 0.0650
Final error: 0.0350
Image:  341
Initial error: 0.0539
Final error: 0.0156
Image:  342
Initial error: 0.0597
Final error: 0.0307
Image:  343
Initial error: 0.0537
Final error: 0.0306
Image:  344
Initial error: 0.1059
Final error: 0.0934
Image:  345
Initial error: 0.0989
Final error: 0.0354
Image:  346
Initial error: 0.0600
Final error: 0.0308
Image:  347
Initial error: 0.0803
Final error: 0.0284
Image:  348
Initial error: 0.0658
Final error: 0.0295
Image:  349
Initial error: 0.0921
Final error: 0.0527
Image:  350
Initial error: 0.1167
Final error: 0.0224
Image:  351
Initial error: 0.0484
Final error: 0.0273
Image:  352
Initial error: 0.2006
Final error: 0.0452
Image:  353
Initial error: 0.1052
Final error: 0.0627
Image:  354
Initial error: 0.1507
Final error: 0.1618
Image:  355
Initial error: 0.0741
Final error: 0.0488
Image:  356
Initial error: 0.0456
Final error: 0.0316
Image:  357
Initial error: 0.1407
Final error: 0.0224
Image:  358
Initial error: 0.0921
Final error: 0.0291
Image:  359
Initial error: 0.0694
Final error: 0.0318
Image:  360
Initial error: 0.0620
Final error: 0.0319
Image:  361
Initial error: 0.2014
Final error: 0.1755
Image:  362
Initial error: 0.0626
Final error: 0.0420
Image:  363
Initial error: 0.0518
Final error: 0.0371
Image:  364
Initial error: 0.1218
Final error: 0.0288
Image:  365
Initial error: 0.1719
Final error: 0.0313
Image:  366
Initial error: 0.0540
Final error: 0.0480
Image:  367
Initial error: 0.0535
Final error: 0.0265
Image:  368
Initial error: 0.0548
Final error: 0.0470
Image:  369
Initial error: 0.0500
Final error: 0.0380
Image:  370
Initial error: 0.0747
Final error: 0.0376
Image:  371
Initial error: 0.1186
Final error: 0.0525
Image:  372
Initial error: 0.0511
Final error: 0.0545
Image:  373
Initial error: 0.1381
Final error: 0.0592
Image:  374
Initial error: 0.1155
Final error: 0.0493
Image:  375
Initial error: 0.0656
Final error: 0.0370
Image:  376
Initial error: 0.0362
Final error: 0.0267
Image:  377
Initial error: 0.0796
Final error: 0.0177
Image:  378
Initial error: 0.0965
Final error: 0.0514
Image:  379
Initial error: 0.0325
Final error: 0.0243
Image:  380
Initial error: 0.1028
Final error: 0.0332
Image:  381
Initial error: 0.0669
Final error: 0.0184
Image:  382
Initial error: 0.1142
Final error: 0.0179
Image:  383
Initial error: 0.0923
Final error: 0.0260
Image:  384
Initial error: 0.1341
Final error: 0.0624
Image:  385
Initial error: 0.0991
Final error: 0.0293
Image:  386
Initial error: 0.0708
Final error: 0.0348
Image:  387
Initial error: 0.0808
Final error: 0.0504
Image:  388
Initial error: 0.0820
Final error: 0.0310
Image:  389
Initial error: 0.0517
Final error: 0.0235
Image:  390
Initial error: 0.0833
Final error: 0.0477
Image:  391
Initial error: 0.1057
Final error: 0.0257
Image:  392
Initial error: 0.1076
Final error: 0.0430
Image:  393
Initial error: 0.0833
Final error: 0.0217
Image:  394
Initial error: 0.0985
Final error: 0.0294
Image:  395
Initial error: 0.2367
Final error: 0.2009
Image:  396
Initial error: 0.1140
Final error: 0.0228
Image:  397
Initial error: 0.0764
Final error: 0.0223
Image:  398
Initial error: 0.0722
Final error: 0.0586
Image:  399
Initial error: 0.1054
Final error: 0.0307
Image:  400
Initial error: 0.0707
Final error: 0.0246
Image:  401
Initial error: 0.0668
Final error: 0.0261
Image:  402
Initial error: 0.0947
Final error: 0.0637
Image:  403
Initial error: 0.1056
Final error: 0.0531
Image:  404
Initial error: 0.0834
Final error: 0.0671
Image:  405
Initial error: 0.0854
Final error: 0.0467
Image:  406
Initial error: 0.0551
Final error: 0.0307
Image:  407
Initial error: 0.0728
Final error: 0.0253
Image:  408
Initial error: 0.1235
Final error: 0.0718
Image:  409
Initial error: 0.0986
Final error: 0.0358
Image:  410
Initial error: 0.0460
Final error: 0.0255
Image:  411
Initial error: 0.0991
Final error: 0.0505
Image:  412
Initial error: 0.0797
Final error: 0.0516
Image:  413
Initial error: 0.1649
Final error: 0.0311
Image:  414
Initial error: 0.1265
Final error: 0.0356
Image:  415
Initial error: 0.1256
Final error: 0.0463
Image:  416
Initial error: 0.0618
Final error: 0.0307
Image:  417
Initial error: 0.0802
Final error: 0.0257
Image:  418
Initial error: 0.1200
Final error: 0.0396
Image:  419
Initial error: 0.1151
Final error: 0.0352
Image:  420
Initial error: 0.0941
Final error: 0.0273
Image:  421
Initial error: 0.0351
Final error: 0.0242
Image:  422
Initial error: 0.1277
Final error: 0.0257
Image:  423
Initial error: 0.0711
Final error: 0.0379
Image:  424
Initial error: 0.0544
Final error: 0.0547
Image:  425
Initial error: 0.0748
Final error: 0.0304
Image:  426
Initial error: 0.0720
Final error: 0.0634
Image:  427
Initial error: 0.0534
Final error: 0.0201
Image:  428
Initial error: 0.0944
Final error: 0.0363
Image:  429
Initial error: 0.1442
Final error: 0.0426
Image:  430
Initial error: 0.0964
Final error: 0.0206
Image:  431
Initial error: 0.0454
Final error: 0.0503
Image:  432
Initial error: 0.0835
Final error: 0.0223
Image:  433
Initial error: 0.1097
Final error: 0.0417
Image:  434
Initial error: 0.0814
Final error: 0.0377
Image:  435
Initial error: 0.1174
Final error: 0.0465
Image:  436
Initial error: 0.1477
Final error: 0.0336
Image:  437
Initial error: 0.1028
Final error: 0.0323
Image:  438
Initial error: 0.0591
Final error: 0.0289
Image:  439
Initial error: 0.0862
Final error: 0.0258
Image:  440
Initial error: 0.1783
Final error: 0.0924
Image:  441
Initial error: 0.0798
Final error: 0.0391
Image:  442
Initial error: 0.0752
Final error: 0.0411
Image:  443
Initial error: 0.0550
Final error: 0.0233
Image:  444
Initial error: 0.0726
Final error: 0.0283
Image:  445
Initial error: 0.1093
Final error: 0.0475
Image:  446
Initial error: 0.1518
Final error: 0.0766
Image:  447
Initial error: 0.0413
Final error: 0.0250
Image:  448
Initial error: 0.1061
Final error: 0.0205
Image:  449
Initial error: 0.0975
Final error: 0.0428
Image:  450
Initial error: 0.1098
Final error: 0.0209
Image:  451
Initial error: 0.1183
Final error: 0.0461
Image:  452
Initial error: 0.1048
Final error: 0.0558
Image:  453
Initial error: 0.0832
Final error: 0.0284
Image:  454
Initial error: 0.0677
Final error: 0.0392
Image:  455
Initial error: 0.1411
Final error: 0.1253
Image:  456
Initial error: 0.0604
Final error: 0.0261
Image:  457
Initial error: 0.0460
Final error: 0.0476
Image:  458
Initial error: 0.0678
Final error: 0.0270
Image:  459
Initial error: 0.1464
Final error: 0.0300
Image:  460
Initial error: 0.0678
Final error: 0.0222
Image:  461
Initial error: 0.1022
Final error: 0.0264
Image:  462
Initial error: 0.0524
Final error: 0.0443
Image:  463
Initial error: 0.1328
Final error: 0.0360
Image:  464
Initial error: 0.0723
Final error: 0.0442
Image:  465
Initial error: 0.1195
Final error: 0.0966
Image:  466
Initial error: 0.0531
Final error: 0.0288
Image:  467
Initial error: 0.0612
Final error: 0.0569
Image:  468
Initial error: 0.0657
Final error: 0.0301
Image:  469
Initial error: 0.0580
Final error: 0.0261
Image:  470
Initial error: 0.0975
Final error: 0.0432
Image:  471
Initial error: 0.1528
Final error: 0.0825
Image:  472
Initial error: 0.1286
Final error: 0.0346
Image:  473
Initial error: 0.0395
Final error: 0.0197
Image:  474
Initial error: 0.0824
Final error: 0.0338
Image:  475
Initial error: 0.0837
Final error: 0.0193
Image:  476
Initial error: 0.1259
Final error: 0.0420
Image:  477
Initial error: 0.1234
Final error: 0.0591
Image:  478
Initial error: 0.0774
Final error: 0.0216
Image:  479
Initial error: 0.0645
Final error: 0.0220
Image:  480
Initial error: 0.0331
Final error: 0.0215
Image:  481
Initial error: 0.0900
Final error: 0.0313
Image:  482
Initial error: 0.0747
Final error: 0.0244
Image:  483
Initial error: 0.1539
Final error: 0.0370
Image:  484
Initial error: 0.0653
Final error: 0.0185
Image:  485
Initial error: 0.0441
Final error: 0.0526
Image:  486
Initial error: 0.1087
Final error: 0.0737
Image:  487
Initial error: 0.1048
Final error: 0.0362
Image:  488
Initial error: 0.1061
Final error: 0.0365
Image:  489
Initial error: 0.0955
Final error: 0.0370
Image:  490
Initial error: 0.1140
Final error: 0.0271
Image:  491
Initial error: 0.0656
Final error: 0.0189
Image:  492
Initial error: 0.1318
Final error: 0.0388
Image:  493
Initial error: 0.0535
Final error: 0.0297
Image:  494
Initial error: 0.1124
Final error: 0.0469
Image:  495
Initial error: 0.0624
Final error: 0.0278
Image:  496
Initial error: 0.2569
Final error: 0.1500
Image:  497
Initial error: 0.1096
Final error: 0.0393
Image:  498
Initial error: 0.0572
Final error: 0.0397
Image:  499
Initial error: 0.0774
Final error: 0.0269
Image:  500
Initial error: 0.0876
Final error: 0.0407
Image:  501
Initial error: 0.0680
Final error: 0.0706
Image:  502
Initial error: 0.0600
Final error: 0.0265
Image:  503
Initial error: 0.0431
Final error: 0.0344
Image:  504
Initial error: 0.1788
Final error: 0.0560
Image:  505
Initial error: 0.0590
Final error: 0.0255
Image:  506
Initial error: 0.0460
Final error: 0.0247

Visualize Results





In [8]:



    


np.mean([fr.final_error(error_type='rmse') for fr in fitter_results])



    


















    
Out[8]:








3.7965750064465333

















In [ ]:



    


from menpofit.visualize import visualize_fitting_results
    
visualize_fitting_results(fitter_results)

PIC





In [9]:



    


sampling_mask = np.require(np.zeros(aam.parts_shape), dtype=np.bool)
sampling_mask[3::7, 3::7] = True

imshow(sampling_mask)



    


















    
Out[9]:








<matplotlib.image.AxesImage at 0x7fcdb6428190>










    
























In [12]:



    


from alabortcvpr2015.aam import PartsAAMFitter, PIC
from alabortcvpr2015.utils import pickle_dump
from alabortcvpr2015.result import SerializableResult

fitter = PartsAAMFitter(aam, algorithm_cls=PIC, n_shape=[3, 12],
                        n_appearance=[25, 50], sampling_mask=sampling_mask)

fitter_results = []

for seed in xrange(repeat):

    np.random.seed(seed=seed)

    for j, i in enumerate(test_images):

        gt_s = i.landmarks[group].lms
        s = fitter.perturb_shape(gt_s, noise_std=0.05)

        fr = fitter.fit(i, s, gt_shape=gt_s, max_iters=20, prior=False)

        fitter_results.append(fr)
        fr.downscale = 0.5

        print 'Image: ', j
        print fr

results = [SerializableResult('none', fr.shapes(), fr.n_iters, 'PIC', fr.gt_shape) 
           for fr in fitter_results]

pickle_dump(results, path + 'PhD/Results/aam_pic_cofw_fast_dsift')



    


















    






Image:  0
Initial error: 0.1455
Final error: 0.0353
Image:  1
Initial error: 0.1943
Final error: 0.1424
Image:  2
Initial error: 0.1043
Final error: 0.0394
Image:  3
Initial error: 0.0740
Final error: 0.0459
Image:  4
Initial error: 0.0568
Final error: 0.0707
Image:  5
Initial error: 0.0896
Final error: 0.0464
Image:  6
Initial error: 0.0515
Final error: 0.0798
Image:  7
Initial error: 0.0844
Final error: 0.0890
Image:  8
Initial error: 0.0743
Final error: 0.0333
Image:  9
Initial error: 0.0349
Final error: 0.0315
Image:  10
Initial error: 0.1329
Final error: 0.0512
Image:  11
Initial error: 0.0988
Final error: 0.0602
Image:  12
Initial error: 0.1064
Final error: 0.1539
Image:  13
Initial error: 0.0610
Final error: 0.1095
Image:  14
Initial error: 0.0597
Final error: 0.0401
Image:  15
Initial error: 0.1343
Final error: 0.0537
Image:  16
Initial error: 0.0953
Final error: 0.0400
Image:  17
Initial error: 0.0350
Final error: 0.0220
Image:  18
Initial error: 0.0438
Final error: 0.0396
Image:  19
Initial error: 0.0525
Final error: 0.0364
Image:  20
Initial error: 0.1728
Final error: 0.1182
Image:  21
Initial error: 0.0794
Final error: 0.0196
Image:  22
Initial error: 0.0570
Final error: 0.0324
Image:  23
Initial error: 0.0569
Final error: 0.0935
Image:  24
Initial error: 0.1400
Final error: 0.0344
Image:  25
Initial error: 0.0925
Final error: 0.0722
Image:  26
Initial error: 0.0780
Final error: 0.0431
Image:  27
Initial error: 0.1338
Final error: 0.0972
Image:  28
Initial error: 0.0598
Final error: 0.0336
Image:  29
Initial error: 0.0710
Final error: 0.0248
Image:  30
Initial error: 0.0756
Final error: 0.1182
Image:  31
Initial error: 0.0886
Final error: 0.0673
Image:  32
Initial error: 0.1199
Final error: 0.0323
Image:  33
Initial error: 0.0646
Final error: 0.0236
Image:  34
Initial error: 0.0666
Final error: 0.0398
Image:  35
Initial error: 0.0633
Final error: 0.0454
Image:  36
Initial error: 0.1225
Final error: 0.0506
Image:  37
Initial error: 0.1070
Final error: 0.0288
Image:  38
Initial error: 0.2081
Final error: 0.0844
Image:  39
Initial error: 0.1125
Final error: 0.0765
Image:  40
Initial error: 0.1040
Final error: 0.0283
Image:  41
Initial error: 0.1221
Final error: 0.0332
Image:  42
Initial error: 0.0702
Final error: 0.0264
Image:  43
Initial error: 0.0794
Final error: 0.0570
Image:  44
Initial error: 0.0574
Final error: 0.0623
Image:  45
Initial error: 0.1360
Final error: 0.0816
Image:  46
Initial error: 0.0614
Final error: 0.0236
Image:  47
Initial error: 0.1019
Final error: 0.0244
Image:  48
Initial error: 0.0547
Final error: 0.0548
Image:  49
Initial error: 0.1382
Final error: 0.0370
Image:  50
Initial error: 0.0935
Final error: 0.0562
Image:  51
Initial error: 0.0507
Final error: 0.0318
Image:  52
Initial error: 0.0813
Final error: 0.0680
Image:  53
Initial error: 0.0541
Final error: 0.0237
Image:  54
Initial error: 0.1199
Final error: 0.1440
Image:  55
Initial error: 0.0797
Final error: 0.0218
Image:  56
Initial error: 0.0550
Final error: 0.0831
Image:  57
Initial error: 0.2140
Final error: 0.1106
Image:  58
Initial error: 0.0827
Final error: 0.0236
Image:  59
Initial error: 0.2586
Final error: 0.1447
Image:  60
Initial error: 0.0609
Final error: 0.0726
Image:  61
Initial error: 0.0739
Final error: 0.0522
Image:  62
Initial error: 0.0645
Final error: 0.0451
Image:  63
Initial error: 0.1605
Final error: 0.1350
Image:  64
Initial error: 0.1035
Final error: 0.0371
Image:  65
Initial error: 0.1046
Final error: 0.0888
Image:  66
Initial error: 0.0749
Final error: 0.0371
Image:  67
Initial error: 0.1410
Final error: 0.0560
Image:  68
Initial error: 0.1158
Final error: 0.0483
Image:  69
Initial error: 0.1469
Final error: 0.0242
Image:  70
Initial error: 0.0782
Final error: 0.0240
Image:  71
Initial error: 0.0872
Final error: 0.0340
Image:  72
Initial error: 0.0511
Final error: 0.0668
Image:  73
Initial error: 0.0904
Final error: 0.0492
Image:  74
Initial error: 0.0529
Final error: 0.0427
Image:  75
Initial error: 0.0771
Final error: 0.0343
Image:  76
Initial error: 0.1009
Final error: 0.0308
Image:  77
Initial error: 0.0674
Final error: 0.0298
Image:  78
Initial error: 0.0708
Final error: 0.0542
Image:  79
Initial error: 0.0310
Final error: 0.0293
Image:  80
Initial error: 0.0736
Final error: 0.0544
Image:  81
Initial error: 0.1060
Final error: 0.0340
Image:  82
Initial error: 0.0856
Final error: 0.0820
Image:  83
Initial error: 0.1551
Final error: 0.0294
Image:  84
Initial error: 0.0984
Final error: 0.0245
Image:  85
Initial error: 0.0657
Final error: 0.0367
Image:  86
Initial error: 0.0528
Final error: 0.0283
Image:  87
Initial error: 0.0888
Final error: 0.0425
Image:  88
Initial error: 0.0645
Final error: 0.0302
Image:  89
Initial error: 0.0694
Final error: 0.0321
Image:  90
Initial error: 0.1102
Final error: 0.0330
Image:  91
Initial error: 0.0489
Final error: 0.0836
Image:  92
Initial error: 0.0618
Final error: 0.0556
Image:  93
Initial error: 0.0567
Final error: 0.0511
Image:  94
Initial error: 0.0565
Final error: 0.0319
Image:  95
Initial error: 0.1755
Final error: 0.0324
Image:  96
Initial error: 0.0939
Final error: 0.0693
Image:  97
Initial error: 0.0607
Final error: 0.0403
Image:  98
Initial error: 0.0642
Final error: 0.0337
Image:  99
Initial error: 0.1066
Final error: 0.1123
Image:  100
Initial error: 0.0566
Final error: 0.0444
Image:  101
Initial error: 0.0460
Final error: 0.0274
Image:  102
Initial error: 0.1414
Final error: 0.0960
Image:  103
Initial error: 0.0969
Final error: 0.0626
Image:  104
Initial error: 0.1130
Final error: 0.0324
Image:  105
Initial error: 0.1171
Final error: 0.0802
Image:  106
Initial error: 0.1626
Final error: 0.0392
Image:  107
Initial error: 0.0946
Final error: 0.0456
Image:  108
Initial error: 0.0640
Final error: 0.0338
Image:  109
Initial error: 0.0630
Final error: 0.0507
Image:  110
Initial error: 0.0563
Final error: 0.0647
Image:  111
Initial error: 0.0975
Final error: 0.0746
Image:  112
Initial error: 0.0880
Final error: 0.0760
Image:  113
Initial error: 0.0977
Final error: 0.0380
Image:  114
Initial error: 0.0317
Final error: 0.0446
Image:  115
Initial error: 0.0545
Final error: 0.0674
Image:  116
Initial error: 0.0892
Final error: 0.1188
Image:  117
Initial error: 0.1502
Final error: 0.0345
Image:  118
Initial error: 0.1312
Final error: 0.0621
Image:  119
Initial error: 0.0860
Final error: 0.0732
Image:  120
Initial error: 0.0807
Final error: 0.0398
Image:  121
Initial error: 0.1010
Final error: 0.0814
Image:  122
Initial error: 0.0853
Final error: 0.0365
Image:  123
Initial error: 0.1709
Final error: 0.0224
Image:  124
Initial error: 0.1103
Final error: 0.0569
Image:  125
Initial error: 0.1276
Final error: 0.1086
Image:  126
Initial error: 0.1330
Final error: 0.0249
Image:  127
Initial error: 0.0568
Final error: 0.0493
Image:  128
Initial error: 0.0784
Final error: 0.0567
Image:  129
Initial error: 0.0825
Final error: 0.0723
Image:  130
Initial error: 0.1277
Final error: 0.0626
Image:  131
Initial error: 0.0869
Final error: 0.0442
Image:  132
Initial error: 0.1745
Final error: 0.1092
Image:  133
Initial error: 0.0741
Final error: 0.1157
Image:  134
Initial error: 0.0380
Final error: 0.0227
Image:  135
Initial error: 0.0680
Final error: 0.0342
Image:  136
Initial error: 0.1427
Final error: 0.0840
Image:  137
Initial error: 0.0531
Final error: 0.0397
Image:  138
Initial error: 0.0824
Final error: 0.0333
Image:  139
Initial error: 0.1574
Final error: 0.0757
Image:  140
Initial error: 0.0859
Final error: 0.0489
Image:  141
Initial error: 0.1135
Final error: 0.0521
Image:  142
Initial error: 0.0967
Final error: 0.0473
Image:  143
Initial error: 0.0709
Final error: 0.0396
Image:  144
Initial error: 0.0746
Final error: 0.0502
Image:  145
Initial error: 0.0995
Final error: 0.0301
Image:  146
Initial error: 0.1562
Final error: 0.0502
Image:  147
Initial error: 0.0296
Final error: 0.0200
Image:  148
Initial error: 0.0746
Final error: 0.0593
Image:  149
Initial error: 0.0788
Final error: 0.0392
Image:  150
Initial error: 0.0865
Final error: 0.0251
Image:  151
Initial error: 0.1486
Final error: 0.0660
Image:  152
Initial error: 0.1261
Final error: 0.0699
Image:  153
Initial error: 0.0593
Final error: 0.1780
Image:  154
Initial error: 0.1148
Final error: 0.0749
Image:  155
Initial error: 0.0424
Final error: 0.0236
Image:  156
Initial error: 0.0696
Final error: 0.0231
Image:  157
Initial error: 0.0897
Final error: 0.0767
Image:  158
Initial error: 0.0709
Final error: 0.0494
Image:  159
Initial error: 0.0770
Final error: 0.0203
Image:  160
Initial error: 0.0896
Final error: 0.0336
Image:  161
Initial error: 0.0566
Final error: 0.0794
Image:  162
Initial error: 0.1261
Final error: 0.0200
Image:  163
Initial error: 0.0371
Final error: 0.0383
Image:  164
Initial error: 0.0896
Final error: 0.0687
Image:  165
Initial error: 0.1579
Final error: 0.1055
Image:  166
Initial error: 0.0461
Final error: 0.0244
Image:  167
Initial error: 0.1395
Final error: 0.0297
Image:  168
Initial error: 0.0635
Final error: 0.0403
Image:  169
Initial error: 0.0577
Final error: 0.0304
Image:  170
Initial error: 0.0677
Final error: 0.0269
Image:  171
Initial error: 0.1362
Final error: 0.1308
Image:  172
Initial error: 0.0548
Final error: 0.0441
Image:  173
Initial error: 0.0878
Final error: 0.0369
Image:  174
Initial error: 0.0631
Final error: 0.0290
Image:  175
Initial error: 0.1121
Final error: 0.0712
Image:  176
Initial error: 0.0838
Final error: 0.0247
Image:  177
Initial error: 0.0726
Final error: 0.0429
Image:  178
Initial error: 0.0756
Final error: 0.1015
Image:  179
Initial error: 0.1107
Final error: 0.0421
Image:  180
Initial error: 0.0490
Final error: 0.0828
Image:  181
Initial error: 0.0922
Final error: 0.0381
Image:  182
Initial error: 0.0570
Final error: 0.0428
Image:  183
Initial error: 0.2303
Final error: 0.2697
Image:  184
Initial error: 0.0729
Final error: 0.0368
Image:  185
Initial error: 0.0955
Final error: 0.0751
Image:  186
Initial error: 0.0428
Final error: 0.0346
Image:  187
Initial error: 0.1089
Final error: 0.0515
Image:  188
Initial error: 0.0475
Final error: 0.0257
Image:  189
Initial error: 0.0721
Final error: 0.0326
Image:  190
Initial error: 0.0402
Final error: 0.0271
Image:  191
Initial error: 0.1259
Final error: 0.0452
Image:  192
Initial error: 0.0933
Final error: 0.0425
Image:  193
Initial error: 0.1342
Final error: 0.0400
Image:  194
Initial error: 0.0999
Final error: 0.0455
Image:  195
Initial error: 0.0704
Final error: 0.0375
Image:  196
Initial error: 0.1382
Final error: 0.0548
Image:  197
Initial error: 0.0759
Final error: 0.0361
Image:  198
Initial error: 0.0925
Final error: 0.0407
Image:  199
Initial error: 0.2003
Final error: 0.1566
Image:  200
Initial error: 0.0296
Final error: 0.0501
Image:  201
Initial error: 0.0662
Final error: 0.0302
Image:  202
Initial error: 0.1082
Final error: 0.0657
Image:  203
Initial error: 0.1051
Final error: 0.0972
Image:  204
Initial error: 0.0694
Final error: 0.0258
Image:  205
Initial error: 0.0642
Final error: 0.0629
Image:  206
Initial error: 0.1591
Final error: 0.0590
Image:  207
Initial error: 0.0743
Final error: 0.0311
Image:  208
Initial error: 0.0834
Final error: 0.0907
Image:  209
Initial error: 0.0302
Final error: 0.0422
Image:  210
Initial error: 0.1182
Final error: 0.0523
Image:  211
Initial error: 0.1392
Final error: 0.1078
Image:  212
Initial error: 0.1727
Final error: 0.1463
Image:  213
Initial error: 0.0967
Final error: 0.0543
Image:  214
Initial error: 0.0693
Final error: 0.0420
Image:  215
Initial error: 0.0467
Final error: 0.0692
Image:  216
Initial error: 0.0462
Final error: 0.0329
Image:  217
Initial error: 0.0689
Final error: 0.0270
Image:  218
Initial error: 0.0753
Final error: 0.0228
Image:  219
Initial error: 0.0913
Final error: 0.0255
Image:  220
Initial error: 0.0454
Final error: 0.0484
Image:  221
Initial error: 0.0881
Final error: 0.0395
Image:  222
Initial error: 0.0522
Final error: 0.0366
Image:  223
Initial error: 0.1234
Final error: 0.0568
Image:  224
Initial error: 0.1384
Final error: 0.0542
Image:  225
Initial error: 0.0403
Final error: 0.0919
Image:  226
Initial error: 0.0928
Final error: 0.0381
Image:  227
Initial error: 0.0977
Final error: 0.0574
Image:  228
Initial error: 0.0862
Final error: 0.0285
Image:  229
Initial error: 0.0313
Final error: 0.0392
Image:  230
Initial error: 0.0766
Final error: 0.0642
Image:  231
Initial error: 0.0461
Final error: 0.0233
Image:  232
Initial error: 0.0966
Final error: 0.0541
Image:  233
Initial error: 0.0930
Final error: 0.0596
Image:  234
Initial error: 0.0836
Final error: 0.0199
Image:  235
Initial error: 0.1278
Final error: 0.0581
Image:  236
Initial error: 0.0792
Final error: 0.0262
Image:  237
Initial error: 0.0720
Final error: 0.0983
Image:  238
Initial error: 0.0644
Final error: 0.0336
Image:  239
Initial error: 0.1429
Final error: 0.0319
Image:  240
Initial error: 0.0946
Final error: 0.0239
Image:  241
Initial error: 0.0618
Final error: 0.0290
Image:  242
Initial error: 0.0744
Final error: 0.0336
Image:  243
Initial error: 0.1122
Final error: 0.0257
Image:  244
Initial error: 0.0575
Final error: 0.0398
Image:  245
Initial error: 0.0544
Final error: 0.0885
Image:  246
Initial error: 0.0538
Final error: 0.0470
Image:  247
Initial error: 0.0780
Final error: 0.0445
Image:  248
Initial error: 0.1808
Final error: 0.0761
Image:  249
Initial error: 0.1456
Final error: 0.1029
Image:  250
Initial error: 0.0451
Final error: 0.0465
Image:  251
Initial error: 0.1166
Final error: 0.0355
Image:  252
Initial error: 0.0973
Final error: 0.0253
Image:  253
Initial error: 0.0885
Final error: 0.0325
Image:  254
Initial error: 0.0466
Final error: 0.0333
Image:  255
Initial error: 0.0679
Final error: 0.0351
Image:  256
Initial error: 0.0830
Final error: 0.0650
Image:  257
Initial error: 0.0432
Final error: 0.0221
Image:  258
Initial error: 0.0673
Final error: 0.0322
Image:  259
Initial error: 0.0865
Final error: 0.0302
Image:  260
Initial error: 0.0908
Final error: 0.0234
Image:  261
Initial error: 0.0740
Final error: 0.0934
Image:  262
Initial error: 0.0900
Final error: 0.0759
Image:  263
Initial error: 0.1658
Final error: 0.0878
Image:  264
Initial error: 0.0854
Final error: 0.1035
Image:  265
Initial error: 0.0829
Final error: 0.0209
Image:  266
Initial error: 0.0311
Final error: 0.0262
Image:  267
Initial error: 0.1021
Final error: 0.0449
Image:  268
Initial error: 0.0379
Final error: 0.0564
Image:  269
Initial error: 0.0634
Final error: 0.1074
Image:  270
Initial error: 0.0793
Final error: 0.0269
Image:  271
Initial error: 0.0943
Final error: 0.2170
Image:  272
Initial error: 0.0656
Final error: 0.0328
Image:  273
Initial error: 0.1027
Final error: 0.0584
Image:  274
Initial error: 0.0564
Final error: 0.0551
Image:  275
Initial error: 0.1014
Final error: 0.0482
Image:  276
Initial error: 0.0445
Final error: 0.0317
Image:  277
Initial error: 0.0804
Final error: 0.0938
Image:  278
Initial error: 0.1770
Final error: 0.0839
Image:  279
Initial error: 0.0711
Final error: 0.0277
Image:  280
Initial error: 0.0745
Final error: 0.0248
Image:  281
Initial error: 0.0704
Final error: 0.0329
Image:  282
Initial error: 0.0894
Final error: 0.0319
Image:  283
Initial error: 0.1841
Final error: 0.0405
Image:  284
Initial error: 0.1010
Final error: 0.0371
Image:  285
Initial error: 0.1247
Final error: 0.0861
Image:  286
Initial error: 0.1321
Final error: 0.1031
Image:  287
Initial error: 0.0635
Final error: 0.0755
Image:  288
Initial error: 0.0757
Final error: 0.0500
Image:  289
Initial error: 0.1199
Final error: 0.0612
Image:  290
Initial error: 0.0550
Final error: 0.0224
Image:  291
Initial error: 0.1876
Final error: 0.0691
Image:  292
Initial error: 0.0652
Final error: 0.0407
Image:  293
Initial error: 0.1901
Final error: 0.1909
Image:  294
Initial error: 0.1188
Final error: 0.0676
Image:  295
Initial error: 0.0774
Final error: 0.0388
Image:  296
Initial error: 0.1737
Final error: 0.0881
Image:  297
Initial error: 0.0759
Final error: 0.0345
Image:  298
Initial error: 0.1079
Final error: 0.1394
Image:  299
Initial error: 0.1248
Final error: 0.0269
Image:  300
Initial error: 0.0788
Final error: 0.0341
Image:  301
Initial error: 0.0819
Final error: 0.0529
Image:  302
Initial error: 0.0746
Final error: 0.0271
Image:  303
Initial error: 0.1413
Final error: 0.0790
Image:  304
Initial error: 0.0938
Final error: 0.0358
Image:  305
Initial error: 0.0980
Final error: 0.0265
Image:  306
Initial error: 0.0504
Final error: 0.0381
Image:  307
Initial error: 0.0532
Final error: 0.0518
Image:  308
Initial error: 0.1361
Final error: 0.0508
Image:  309
Initial error: 0.0621
Final error: 0.0263
Image:  310
Initial error: 0.0880
Final error: 0.0364
Image:  311
Initial error: 0.1025
Final error: 0.0479
Image:  312
Initial error: 0.0863
Final error: 0.0282
Image:  313
Initial error: 0.1403
Final error: 0.0686
Image:  314
Initial error: 0.1173
Final error: 0.0400
Image:  315
Initial error: 0.0981
Final error: 0.0422
Image:  316
Initial error: 0.1342
Final error: 0.0302
Image:  317
Initial error: 0.1548
Final error: 0.0527
Image:  318
Initial error: 0.0686
Final error: 0.0503
Image:  319
Initial error: 0.0933
Final error: 0.0290
Image:  320
Initial error: 0.0571
Final error: 0.0988
Image:  321
Initial error: 0.0915
Final error: 0.0777
Image:  322
Initial error: 0.1291
Final error: 0.0494
Image:  323
Initial error: 0.1200
Final error: 0.1061
Image:  324
Initial error: 0.1173
Final error: 0.0635
Image:  325
Initial error: 0.0479
Final error: 0.0228
Image:  326
Initial error: 0.0901
Final error: 0.0904
Image:  327
Initial error: 0.0582
Final error: 0.0281
Image:  328
Initial error: 0.0629
Final error: 0.0285
Image:  329
Initial error: 0.1068
Final error: 0.0874
Image:  330
Initial error: 0.0460
Final error: 0.0280
Image:  331
Initial error: 0.0775
Final error: 0.0899
Image:  332
Initial error: 0.1022
Final error: 0.0429
Image:  333
Initial error: 0.1114
Final error: 0.0900
Image:  334
Initial error: 0.0993
Final error: 0.1256
Image:  335
Initial error: 0.0867
Final error: 0.0334
Image:  336
Initial error: 0.1043
Final error: 0.0630
Image:  337
Initial error: 0.1331
Final error: 0.0752
Image:  338
Initial error: 0.1272
Final error: 0.0332
Image:  339
Initial error: 0.1505
Final error: 0.0399
Image:  340
Initial error: 0.0650
Final error: 0.0533
Image:  341
Initial error: 0.0539
Final error: 0.0266
Image:  342
Initial error: 0.0597
Final error: 0.0446
Image:  343
Initial error: 0.0537
Final error: 0.1370
Image:  344
Initial error: 0.1059
Final error: 0.0689
Image:  345
Initial error: 0.0989
Final error: 0.0449
Image:  346
Initial error: 0.0600
Final error: 0.0378
Image:  347
Initial error: 0.0803
Final error: 0.0488
Image:  348
Initial error: 0.0658
Final error: 0.0342
Image:  349
Initial error: 0.0921
Final error: 0.0557
Image:  350
Initial error: 0.1167
Final error: 0.0872
Image:  351
Initial error: 0.0484
Final error: 0.0447
Image:  352
Initial error: 0.2006
Final error: 0.0476
Image:  353
Initial error: 0.1052
Final error: 0.0827
Image:  354
Initial error: 0.1507
Final error: 0.2511
Image:  355
Initial error: 0.0741
Final error: 0.0567
Image:  356
Initial error: 0.0456
Final error: 0.0372
Image:  357
Initial error: 0.1407
Final error: 0.0348
Image:  358
Initial error: 0.0921
Final error: 0.0699
Image:  359
Initial error: 0.0694
Final error: 0.0940
Image:  360
Initial error: 0.0620
Final error: 0.0454
Image:  361
Initial error: 0.2014
Final error: 0.0854
Image:  362
Initial error: 0.0626
Final error: 0.0494
Image:  363
Initial error: 0.0518
Final error: 0.0423
Image:  364
Initial error: 0.1218
Final error: 0.0643
Image:  365
Initial error: 0.1719
Final error: 0.0836
Image:  366
Initial error: 0.0540
Final error: 0.0517
Image:  367
Initial error: 0.0535
Final error: 0.0368
Image:  368
Initial error: 0.0548
Final error: 0.0406
Image:  369
Initial error: 0.0500
Final error: 0.0409
Image:  370
Initial error: 0.0747
Final error: 0.0446
Image:  371
Initial error: 0.1186
Final error: 0.0791
Image:  372
Initial error: 0.0511
Final error: 0.0685
Image:  373
Initial error: 0.1381
Final error: 0.0429
Image:  374
Initial error: 0.1155
Final error: 0.0653
Image:  375
Initial error: 0.0656
Final error: 0.0497
Image:  376
Initial error: 0.0362
Final error: 0.0436
Image:  377
Initial error: 0.0796
Final error: 0.0286
Image:  378
Initial error: 0.0965
Final error: 0.0771
Image:  379
Initial error: 0.0325
Final error: 0.0277
Image:  380
Initial error: 0.1028
Final error: 0.0816
Image:  381
Initial error: 0.0669
Final error: 0.0344
Image:  382
Initial error: 0.1142
Final error: 0.0385
Image:  383
Initial error: 0.0923
Final error: 0.0405
Image:  384
Initial error: 0.1341
Final error: 0.0576
Image:  385
Initial error: 0.0991
Final error: 0.0247
Image:  386
Initial error: 0.0708
Final error: 0.0369
Image:  387
Initial error: 0.0808
Final error: 0.1426
Image:  388
Initial error: 0.0820
Final error: 0.0338
Image:  389
Initial error: 0.0517
Final error: 0.0411
Image:  390
Initial error: 0.0833
Final error: 0.0732
Image:  391
Initial error: 0.1057
Final error: 0.0322
Image:  392
Initial error: 0.1076
Final error: 0.0477
Image:  393
Initial error: 0.0833
Final error: 0.0210
Image:  394
Initial error: 0.0985
Final error: 0.0456
Image:  395
Initial error: 0.2367
Final error: 0.1462
Image:  396
Initial error: 0.1140
Final error: 0.0276
Image:  397
Initial error: 0.0764
Final error: 0.0364
Image:  398
Initial error: 0.0722
Final error: 0.0521
Image:  399
Initial error: 0.1054
Final error: 0.0370
Image:  400
Initial error: 0.0707
Final error: 0.0243
Image:  401
Initial error: 0.0668
Final error: 0.0685
Image:  402
Initial error: 0.0947
Final error: 0.0422
Image:  403
Initial error: 0.1056
Final error: 0.1028
Image:  404
Initial error: 0.0834
Final error: 0.0524
Image:  405
Initial error: 0.0854
Final error: 0.0417
Image:  406
Initial error: 0.0551
Final error: 0.0506
Image:  407
Initial error: 0.0728
Final error: 0.0274
Image:  408
Initial error: 0.1235
Final error: 0.0704
Image:  409
Initial error: 0.0986
Final error: 0.0320
Image:  410
Initial error: 0.0460
Final error: 0.0436
Image:  411
Initial error: 0.0991
Final error: 0.1091
Image:  412
Initial error: 0.0797
Final error: 0.0810
Image:  413
Initial error: 0.1649
Final error: 0.0906
Image:  414
Initial error: 0.1265
Final error: 0.0331
Image:  415
Initial error: 0.1256
Final error: 0.1198
Image:  416
Initial error: 0.0618
Final error: 0.0467
Image:  417
Initial error: 0.0802
Final error: 0.0215
Image:  418
Initial error: 0.1200
Final error: 0.0474
Image:  419
Initial error: 0.1151
Final error: 0.0850
Image:  420
Initial error: 0.0941
Final error: 0.0227
Image:  421
Initial error: 0.0351
Final error: 0.0570
Image:  422
Initial error: 0.1277
Final error: 0.0544
Image:  423
Initial error: 0.0711
Final error: 0.0477
Image:  424
Initial error: 0.0544
Final error: 0.0515
Image:  425
Initial error: 0.0748
Final error: 0.0326
Image:  426
Initial error: 0.0720
Final error: 0.0869
Image:  427
Initial error: 0.0534
Final error: 0.0352
Image:  428
Initial error: 0.0944
Final error: 0.0438
Image:  429
Initial error: 0.1442
Final error: 0.0554
Image:  430
Initial error: 0.0964
Final error: 0.0209
Image:  431
Initial error: 0.0454
Final error: 0.0699
Image:  432
Initial error: 0.0835
Final error: 0.0229
Image:  433
Initial error: 0.1097
Final error: 0.0537
Image:  434
Initial error: 0.0814
Final error: 0.0669
Image:  435
Initial error: 0.1174
Final error: 0.0554
Image:  436
Initial error: 0.1477
Final error: 0.1153
Image:  437
Initial error: 0.1028
Final error: 0.0438
Image:  438
Initial error: 0.0591
Final error: 0.0302
Image:  439
Initial error: 0.0862
Final error: 0.0413
Image:  440
Initial error: 0.1783
Final error: 0.1824
Image:  441
Initial error: 0.0798
Final error: 0.0437
Image:  442
Initial error: 0.0752
Final error: 0.0460
Image:  443
Initial error: 0.0550
Final error: 0.1008
Image:  444
Initial error: 0.0726
Final error: 0.0375
Image:  445
Initial error: 0.1093
Final error: 0.1422
Image:  446
Initial error: 0.1518
Final error: 0.0959
Image:  447
Initial error: 0.0413
Final error: 0.0331
Image:  448
Initial error: 0.1061
Final error: 0.0320
Image:  449
Initial error: 0.0975
Final error: 0.1071
Image:  450
Initial error: 0.1098
Final error: 0.0245
Image:  451
Initial error: 0.1183
Final error: 0.1157
Image:  452
Initial error: 0.1048
Final error: 0.0987
Image:  453
Initial error: 0.0832
Final error: 0.0631
Image:  454
Initial error: 0.0677
Final error: 0.0414
Image:  455
Initial error: 0.1411
Final error: 0.0906
Image:  456
Initial error: 0.0604
Final error: 0.0304
Image:  457
Initial error: 0.0460
Final error: 0.0810
Image:  458
Initial error: 0.0678
Final error: 0.0411
Image:  459
Initial error: 0.1464
Final error: 0.0329
Image:  460
Initial error: 0.0678
Final error: 0.0309
Image:  461
Initial error: 0.1022
Final error: 0.0294
Image:  462
Initial error: 0.0524
Final error: 0.0603
Image:  463
Initial error: 0.1328
Final error: 0.0359
Image:  464
Initial error: 0.0723
Final error: 0.1272
Image:  465
Initial error: 0.1195
Final error: 0.1484
Image:  466
Initial error: 0.0531
Final error: 0.0311
Image:  467
Initial error: 0.0612
Final error: 0.0611
Image:  468
Initial error: 0.0657
Final error: 0.0306
Image:  469
Initial error: 0.0580
Final error: 0.0728
Image:  470
Initial error: 0.0975
Final error: 0.0660
Image:  471
Initial error: 0.1528
Final error: 0.0941
Image:  472
Initial error: 0.1286
Final error: 0.0277
Image:  473
Initial error: 0.0395
Final error: 0.0346
Image:  474
Initial error: 0.0824
Final error: 0.0320
Image:  475
Initial error: 0.0837
Final error: 0.0231
Image:  476
Initial error: 0.1259
Final error: 0.0729
Image:  477
Initial error: 0.1234
Final error: 0.0598
Image:  478
Initial error: 0.0774
Final error: 0.0309
Image:  479
Initial error: 0.0645
Final error: 0.0460
Image:  480
Initial error: 0.0331
Final error: 0.0253
Image:  481
Initial error: 0.0900
Final error: 0.0625
Image:  482
Initial error: 0.0747
Final error: 0.0332
Image:  483
Initial error: 0.1539
Final error: 0.0437
Image:  484
Initial error: 0.0653
Final error: 0.0258
Image:  485
Initial error: 0.0441
Final error: 0.0539
Image:  486
Initial error: 0.1087
Final error: 0.0806
Image:  487
Initial error: 0.1048
Final error: 0.0542
Image:  488
Initial error: 0.1061
Final error: 0.0832
Image:  489
Initial error: 0.0955
Final error: 0.0528
Image:  490
Initial error: 0.1140
Final error: 0.0277
Image:  491
Initial error: 0.0656
Final error: 0.0247
Image:  492
Initial error: 0.1318
Final error: 0.0572
Image:  493
Initial error: 0.0535
Final error: 0.0371
Image:  494
Initial error: 0.1124
Final error: 0.0571
Image:  495
Initial error: 0.0624
Final error: 0.0354
Image:  496
Initial error: 0.2569
Final error: 0.1255
Image:  497
Initial error: 0.1096
Final error: 0.0383
Image:  498
Initial error: 0.0572
Final error: 0.0800
Image:  499
Initial error: 0.0774
Final error: 0.0282
Image:  500
Initial error: 0.0876
Final error: 0.0614
Image:  501
Initial error: 0.0680
Final error: 0.0828
Image:  502
Initial error: 0.0600
Final error: 0.0470
Image:  503
Initial error: 0.0431
Final error: 0.0328
Image:  504
Initial error: 0.1788
Final error: 0.0845
Image:  505
Initial error: 0.0590
Final error: 0.0478
Image:  506
Initial error: 0.0460
Final error: 0.0297

Visualize Results





In [11]:



    


np.mean([fr.final_error(error_type='rmse') for fr in fitter_results])



    


















    
Out[11]:








5.4382533925833316

















In [ ]:



    


from menpofit.visualize import visualize_fitting_results
    
visualize_fitting_results(fitter_results)

Content source: jalabort/alabortcvpr2015

Similar notebooks:

notebook.community | gallery | about