-
Notifications
You must be signed in to change notification settings - Fork 3
/
IJB_evaluation.py
828 lines (718 loc) · 35.4 KB
/
IJB_evaluation.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
# coding: utf-8
import os
import pickle
import matplotlib
import pandas as pd
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import timeit
import sklearn
import argparse
from sklearn.metrics import roc_curve, auc
from menpo.visualize.viewmatplotlib import sample_colours_from_colourmap
from prettytable import PrettyTable
from pathlib import Path
import sys
import warnings
import platform
# import onnx
import math
# from util import utils as utils
from einops import rearrange, repeat
# if 'Alienware' in platform.node():
# os.environ['CUDA_VISIBLE_DEVICES'] = '0'
# else:
# os.environ['CUDA_VISIBLE_DEVICES'] = '3'
sys.path.insert(0, "../")
# sys.path.append('/homes/zs003/projects/paper_face/paper_face/vit_pytorch_my')
warnings.filterwarnings("ignore")
parser = argparse.ArgumentParser(description='do ijb test')
# general
# parser.add_argument('--model-prefix', default='/home/zhonglin/project/paper_face/paper_face/results/49_12vit_30epoch_lr5e4/Backbone_VIT_land_Epoch_29_Batch_454839_Time_2021-07-03-05-44_checkpoint.pth', help='path to load model.')
#/import/nobackup_mmv_ioannisp/zs003/face_rec/pretrain_net/backbone.pth
# parser.add_argument('--image-path', default='/home/zhonglin/mount_folder/dataset/face_rec/IJB/IJB_release/IJBC', type=str, help='')#
#/home/zhonglin/mount_folder/dataset/face_rec/IJB/IJB_release/IJBC
#/import/nobackup_mmv_ioannisp/zs003/face_rec/IJB_release/IJBC
parser.add_argument('--result-dir', default='ms1mv3_arcface_r50', type=str, help='')
parser.add_argument('--batch-size', default=360, type=int, help='')#480
parser.add_argument('--network', default='iresnet50', type=str, help='')
parser.add_argument('--job', default='ms1mv3_arcface_r50', type=str, help='job name')
parser.add_argument('--target', default='IJBB', type=str, help='target, set to IJBC or IJBB')
args = parser.parse_args()
target = args.target
# model_path = args.model_prefix
# image_path = args.image_path
do_onnx=False
os.environ['CUDA_VISIBLE_DEVICES'] = '0'
output_path='/home/zhonglin/project/paper_face/paper_face/onnx_check'
model_path='/data/scratch/acw569/checkpoint/sp_check/webface_webland_34epoch_mixup0501_aug01/Backbone_VIT_land_8_Epoch_34_Batch_359865_Time_2024-09-15-10-19_checkpoint.pth'
image_path='/data/scratch/acw569/face/IJB/IJB_release/IJBB'
result_dir = args.result_dir
gpu_id = None
use_norm_score = True # if Ture, TestMode(N1)
use_detector_score = True # if Ture, TestMode(D1)
use_flip_test = True # if Ture, TestMode(F1)
job = args.job
batch_size = args.batch_size
import cv2
import numpy as np
import torch
from skimage import transform as trans
# import backbones
import pdb
from PIL import Image
# from vit_modify import ViT_stn_land
class Embedding(object):
def __init__(self, prefix, data_shape, batch_size=1):
image_size = (112, 112)
self.image_size = image_size
# pdb.set_trace()
weight = torch.load(prefix,map_location=lambda storage, loc: storage.cuda(0))#vit+my resnet
# weight = torch.load(prefix, map_location='cpu')['state_dict'] #adaface
# pdb.set_trace()
# resnet = eval("backbones.{}".format(args.network))(False).cuda()
# resnet = eval("backbones.{}".format(args.network))(False).cuda()
# resnet=ViT_stn_land(
# image_size = 112,
# patch_size = 16,
# num_classes = 512,
# dim = 512,
# depth = 12,
# heads = 8,
# mlp_dim = 2048,
# dropout=0.1,
# emb_dropout=0.1
# ).cuda()#to(conf.device)#
# pdb.set_trace()
# # from vit_pytorch_my.vits_face import ViT_face_landmark
# from vit_pytorch_my.vit_face import ViT_face,ViT_face_landmark,ViT_face_landmark_patch8,ViT_face_landmark_halfpatchsize, ViT_face_landmark_astoken,ViT_face_landmark_halfpatchsize_globaltoken,ViT_face_landmark_globaltoken,ViT_face_landmark_patch8_global
# from vit_pytorch_my.vit_face import ViT_face_landmark_patch8_overlap,ViT_face_landmark_halfpatchsize_scale_affine,ViT_face_landmark_scale_affine,ViT_face_landmark_patch8_scale_affine,ViT_face_landmark_largepatch_inner,ViT_face_landmark_scale_affine_global,ViT_face_landmark_halfpatchsize_scale_affine_global
# from vit_pytorch_my.vit_face import ViT_face_landmark_patch8_scale_affine_global,ViT_face_landmark_halfpatchsize_testnewloss,ViT_face_landmark_patch8_global_sysgraph,ViT_face_landmark_patch8_global_graph_lessparam_more,ViTs_face_overlap
# from vit_pytorch_my.vit_face import ViT_face_landmark_patch8_att
# # from vit_pytorch_my.cross_vit import cross_VIT,cross_VIT_landmark
from functools import partial
NUM_CLASS=205990#360232,205990,93431, 56000,ir-se-50 12000/10000
from face_pre_pro.ViT_face import ViT_face_landmark_patch8
resnet= ViT_face_landmark_patch8(
loss_type = 'CosFace',
GPU_ID = '0',
num_class = NUM_CLASS,#205990 93431# 56000
image_size=112,
patch_size=8,#8 14
dim=768,#512 ,768
depth=12,#20,12
heads=11,
mlp_dim=2048,
dropout=0.1,
emb_dropout=0.1,
with_land=True
).cuda()
# ####VIT load
# # resnet.load_state_dict(weight,strict=True)
# # model = torch.nn.DataParallel(torch.nn.DataParallel(resnet))
model = torch.nn.DataParallel(resnet)
model.load_state_dict(weight,strict=True)
##adaface load
# print('statedict keys:',statedict.keys())
# print('backbone keys:',backbone.keys())
# backbone.load_state_dict({k.replace("module.", ''): v for k, v in statedict.items() if 'module.' in k},strict=True)
# resnet.load_state_dict({k.replace("module.", ''): v for k, v in weight.items() if 'module.' in k},strict=True)
# resnet.load_state_dict({k.replace("model.", ''): v for k, v in weight.items() if 'model.' in k},strict=False)
# model = torch.nn.DataParallel(resnet)
self.model = model
self.model.eval()
# pdb.set_trace()
if do_onnx:
convert_onnx(model,path_module=model_path,output_path=output_path,simplify=False)
sys.exit()
src = np.array([
[30.2946, 51.6963],
[65.5318, 51.5014],
[48.0252, 71.7366],
[33.5493, 92.3655],
[62.7299, 92.2041]], dtype=np.float32)
src[:, 0] += 8.0
self.src = src
self.batch_size = batch_size
self.data_shape = data_shape
self.pre_land=False
self.keep_land=False
# pdb.set_trace()
if self.pre_land==True:
from vit_pytorch_my.vit_face import face_landmark_4simmin_glo_loc
self.landmarkcnn=face_landmark_4simmin_glo_loc(loss_type = 'CosFace',
GPU_ID = None,
num_class = 30000,
num_patches=196,
image_size=112,
patch_size=8,#8
dim=512,#512
depth=12,#20
heads=11,#8
mlp_dim=2560,
dropout=0.1,
emb_dropout=0.1)
self.landmarkcnn=self.landmarkcnn.cuda()
#/import/nobackup_mmv_ioannisp/zs003/checkpoints/face_rec/ssl_results/simmin_vit_land_real/simmim_pretrain/simmim_pretrain__vit_face_100epo/ckpt_epoch_99.pth
# load_part_checkpoint_landmark_fromsimmim(path='/root/face/check/ckpt_epoch_99.pth',model=landmarkcnn,pretrain_name=['stn','output'])
# load_part_checkpoint_landmark_fromsimmim(path='/import/nobackup_mmv_ioannisp/zs003/checkpoints/face_rec/ssl_results/simmin_vit_land_real_gnn2trans/simmim_pretrain/simmim_pretrain__vit_face_100epo/ckpt_epoch_99.pth',model=landmarkcnn,pretrain_name=['stn','output'])
# load_part_checkpoint_landmark(path='/import/nobackup_mmv_ioannisp/zs003/checkpoints/face_rec/results/VGG_landmark_new144/Backbone_VIT_land_8_Epoch_34_Batch_148113_Time_2022-09-05-20-37_checkpoint.pth',model=self.landmarkcnn,pretrain_name=['stn','output'])
#webface
# load_part_checkpoint_landmark(path='/import/nobackup_mmv_ioannisp/zs003/checkpoints/face_rec/results/webface_196landmark/Backbone_VIT_land_8_Epoch_34_Batch_327225_Time_2022-05-05-10-34_checkpoint.pth',model=self.landmarkcnn,pretrain_name=['stn','output'])
#hpc
# load_part_checkpoint_landmark(path='/data/scratch/acw569/precheckpoint/webface_196land_sp/Backbone_VIT_land_8_Epoch_34_Batch_327225_Time_2022-05-05-10-34_checkpoint.pth',model=self.landmarkcnn,pretrain_name=['stn','output'])
#waixingren
# load_part_checkpoint_landmark(path='/home/zhonglin/mount_folder/dataset/checkpoints/web_196_land/Backbone_VIT_land_8_Epoch_34_Batch_327225_Time_2022-05-05-10-34_checkpoint.pth',model=self.landmarkcnn,pretrain_name=['stn','output'])
#ms1m
# load_part_checkpoint_landmark(path='/import/nobackup_mmv_ioannisp/zs003/checkpoints/face_rec/results/4gpu_landmark_augall_again_stnsmalldecay/Backbone_VIT_land_8_Epoch_34_Batch_523881_Time_2021-07-31-11-07_checkpoint.pth',model=self.landmarkcnn,pretrain_name=['stn','output'])
# load_part_checkpoint_landmark(path='/import/nobackup_mmv_ioannisp/zs003/checkpoints/face_rec/results/webface_196landmark/Backbone_VIT_land_8_Epoch_34_Batch_327225_Time_2022-05-05-10-34_checkpoint.pth',model=self.landmarkcnn,pretrain_name=['stn','output'])
self.landmarkcnn.eval()
# if knowledge_dis:
transf_cit = torch.nn.MSELoss()
def get(self, rimg, landmark,index=0):
assert landmark.shape[0] == 68 or landmark.shape[0] == 5
assert landmark.shape[1] == 2
if landmark.shape[0] == 68:
landmark5 = np.zeros((5, 2), dtype=np.float32)
landmark5[0] = (landmark[36] + landmark[39]) / 2
landmark5[1] = (landmark[42] + landmark[45]) / 2
landmark5[2] = landmark[30]
landmark5[3] = landmark[48]
landmark5[4] = landmark[54]
else:
landmark5 = landmark
tform = trans.SimilarityTransform()
tform.estimate(landmark5, self.src)
M = tform.params[0:2, :]
img = cv2.warpAffine(rimg,
M, (self.image_size[1], self.image_size[0]),
borderValue=0.0)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) #resnet rgb,,mine bgr
# pdb.set_trace()
# index=0
_data_pil = Image.fromarray(img)
# _data_pil.save("./img_check/{}_test_trans_ijb.jpeg".format(index))
img_flip = np.fliplr(img)# keep this one
_data_flip_pil = Image.fromarray(img_flip)
# _data_flip_pil.save("./img_check/{}_test_trans_ijb_flip.jpeg".format(index))
img = np.transpose(img, (2, 0, 1)) # 3*112*112, RGB
img_flip = np.transpose(img_flip, (2, 0, 1))
input_blob = np.zeros((2, 3, self.image_size[1], self.image_size[0]), dtype=np.uint8)
input_blob[0] = img
input_blob[1] = img_flip
return input_blob
@torch.no_grad()
def forward_db(self, batch_data):
imgs = torch.Tensor(batch_data).cuda()
imgs.div_(255).sub_(0.5)#.div_(0.5) #my 0.5, arcface 1 #retina data, 255
# pdb.set_trace()
# if self.pre_land==True:
# imgs=torch.Tensor(imgs).cuda()
# land_label,img_reconstructed=self.landmarkcnn(imgs.float())#div 255/2
# # land_label,img_reconstructed=landmarkcnn(images[0])
# #reconstructed image to embedding
# if not self.keep_land:
# imgs = rearrange(img_reconstructed, 'b c (h p1) (w p2) -> b (h w) (p1 p2 c)', p1 = landmarkcnn.patch_size, p2 = landmarkcnn.patch_size)
# # batch_data=batch_data.cpu().numpy()#np.array(batch_data)
feat = self.model(imgs)
feat = feat.reshape([self.batch_size, 2 * feat.shape[1]])
return feat.cpu().numpy()
@torch.no_grad()
def forward_db_visual(self, batch_data):
# pdb.set_trace()
imgs = torch.Tensor(batch_data).cuda()
imgs.div_(255).sub_(0.5)#.div_(0.5) #my 0.5, arcface 1 #retina data, 255
# feat = self.model(imgs)
if self.pre_land==True:
# imgs=torch.Tensor(imgs).cuda()
land_label,img_reconstructed=self.landmarkcnn(imgs.float())#div 255/2
# land_label,img_reconstructed=landmarkcnn(images[0])
#reconstructed image to embedding
if not self.keep_land:
imgs = rearrange(img_reconstructed, 'b c (h p1) (w p2) -> b (h w) (p1 p2 c)', p1 = self.landmarkcnn.patch_size, p2 = self.landmarkcnn.patch_size)
feat,theta = self.model(imgs,visualize=True)
feat = feat.reshape([self.batch_size, 2 * feat.shape[1]])
return feat.cpu().numpy(),theta#.cpu().numpy()
# 将一个list尽量均分成n份,限制len(list)==n,份数大于原list内元素个数则分配空list[]
def divideIntoNstrand(listTemp, n):
twoList = [[] for i in range(n)]
for i, e in enumerate(listTemp):
twoList[i % n].append(e)
return twoList
def read_template_media_list(path):
# ijb_meta = np.loadtxt(path, dtype=str)
ijb_meta = pd.read_csv(path, sep=' ', header=None).values
templates = ijb_meta[:, 1].astype(np.int)
medias = ijb_meta[:, 2].astype(np.int)
return templates, medias
# In[ ]:
def read_template_pair_list(path):
# pairs = np.loadtxt(path, dtype=str)
pairs = pd.read_csv(path, sep=' ', header=None).values
# print(pairs.shape)
# print(pairs[:, 0].astype(np.int))
t1 = pairs[:, 0].astype(np.int)
t2 = pairs[:, 1].astype(np.int)
label = pairs[:, 2].astype(np.int)
return t1, t2, label
# In[ ]:
def read_image_feature(path):
with open(path, 'rb') as fid:
img_feats = pickle.load(fid)
return img_feats
def calculate_overlap_near(theta,patch_size):
theta=theta.cpu().numpy()
# pdb.set_trace()
out_mean=[]
half_patch_size=np.int16(patch_size*0.5)
for sin_sample in theta:
overlap_map=np.zeros([len(sin_sample)])
for i in range(len(sin_sample)):
sin_theta=sin_sample[i]
X=sin_theta
# X = [0,0] # Your cooridinate
x1 = X[0]
y1= X[1]
array = sin_sample#[[1,1],[0,1],[1,0],[-2,2]] # Sample array of data
smallestDistance = 9999 # Make it big so it can be replaced immediately, essentially just a placeholder
# for point in array:
for j in range(len(array)):
point=array[j]
x2 = point[0]
y2 = point[1]
separation = math.hypot(x2 - x1, y2 - y1) #Distance equation in easy format
if separation < smallestDistance and separation!=0: # Could make this <= instead of < if you want to replace any ties for closest point
smallestDistance = separation
closestPoint = point
smallest_index=j
# pdb.set_trace()
#calculate the overlap
sin_theta=np.int64(np.around(sin_theta))
sin_map=np.zeros([112,112])
x_min_sin=max(0,sin_theta[0]-half_patch_size)
x_max_sin=min(111,sin_theta[0]+half_patch_size)
y_min_sin=max(0,sin_theta[1]-half_patch_size)
y_max_sin=min(111,sin_theta[1]+half_patch_size)
sin_map[x_min_sin:x_max_sin,y_min_sin:y_max_sin]=1
#
right_theta=sin_sample[smallest_index]
right_theta=np.int64(np.around(right_theta))
right_map=np.zeros([112,112])
x_min_rig=max(0,right_theta[0]-half_patch_size)
x_max_rig=min(111,right_theta[0]+half_patch_size)
y_min_rig=max(0,right_theta[1]-half_patch_size)
y_max_rig=min(111,right_theta[1]+half_patch_size)
right_map[x_min_rig:x_max_rig,y_min_rig:y_max_rig]=1
# right_map[right_theta[0]-half_patch_size:right_theta[0]+half_patch_size,right_theta[1]-half_patch_size:right_theta[1]+half_patch_size]=1
out_map=sin_map+right_map
out_index=np.where(out_map==2)
overlap_map[i]=len(out_index[0])/(patch_size*patch_size)
# pdb.set_trace()
one_mean=np.mean(overlap_map)
out_mean+=[one_mean]
# pdb.set_trace()
return out_mean
# for i in range(len(sin_sample)):
# sin_theta=sin_sample[i]
# search_nodes=np.delete(sin_sample.copy(),i,axis=0)
# closest_node(sin_theta,sin_sample)
# for sin_sample in theta:
# half_patch_size=np.int16(patch_size*0.5)
# overlap_map=np.zeros([len(sin_sample),len(sin_sample)])
# for i in range(len(sin_sample)):
# sin_theta=sin_sample[i].cpu().numpy()
# sin_theta=np.int64(np.around(sin_theta))
# sin_map=np.zeros([112,112])
# x_min_sin=max(0,sin_theta[0]-half_patch_size)
# x_max_sin=min(111,sin_theta[0]+half_patch_size)
# y_min_sin=max(0,sin_theta[1]-half_patch_size)
# y_max_sin=min(111,sin_theta[1]+half_patch_size)
# sin_map[x_min_sin:x_max_sin,y_min_sin:y_max_sin]=1
# # sin_map[sin_theta[0]-half_patch_size:sin_theta[0]+half_patch_size,sin_theta[1]-half_patch_size:sin_theta[1]+half_patch_size]=1
# for j in range(len(sin_sample)):
# if i==j:
# continue
# right_theta=sin_sample[j].cpu().numpy()
# right_theta=np.int64(np.around(right_theta))
# right_map=np.zeros([112,112])
# x_min_rig=max(0,right_theta[0]-half_patch_size)
# x_max_rig=min(111,right_theta[0]+half_patch_size)
# y_min_rig=max(0,right_theta[1]-half_patch_size)
# y_max_rig=min(111,right_theta[1]+half_patch_size)
# right_map[x_min_rig:x_max_rig,y_min_rig:y_max_rig]=1
# # right_map[right_theta[0]-half_patch_size:right_theta[0]+half_patch_size,right_theta[1]-half_patch_size:right_theta[1]+half_patch_size]=1
# out_map=sin_map+right_map
# out_index=np.where(out_map==2)
# overlap_map[i,j]=len(out_index[0])/(len(sin_sample)*len(sin_sample))
# pdb.set_trace()
# one_mean=np.mean(overlap_map)
# out_mean+=[one_mean]
# In[ ]:
def get_image_feature(img_path, files_list, model_path, epoch, gpu_id, save_samples=True,overlap=False,logpath='ijb'):
batch_size = args.batch_size
data_shape = (3, 112, 112)
files = files_list
print('files:', len(files))
rare_size = len(files) % batch_size
faceness_scores = []
batch = 0
img_feats = np.empty((len(files), 768*2), dtype=np.float32)#768*3,512*2
batch_data = np.empty((2 * batch_size, 3, 112, 112))
embedding = Embedding(model_path, data_shape, batch_size)
over_lap_all=[]
for img_index, each_line in enumerate(files[:len(files) - rare_size]):
name_lmk_score = each_line.strip().split(' ')
img_name = os.path.join(img_path, name_lmk_score[0])
img = cv2.imread(img_name)
lmk = np.array([float(x) for x in name_lmk_score[1:-1]],
dtype=np.float32)
lmk = lmk.reshape((5, 2))
input_blob = embedding.get(img, lmk,index=img_index)
batch_data[2 * (img_index - batch * batch_size)][:] = input_blob[0]
batch_data[2 * (img_index - batch * batch_size) + 1][:] = input_blob[1]
if (img_index + 1) % batch_size == 0:
print('batch', batch)
# pdb.set_trace()
if not save_samples:
img_feats[batch * batch_size:batch * batch_size +
batch_size][:] = embedding.forward_db(batch_data)
else:
# if (visualize==True) and (overlap==True):
img_feats[batch * batch_size:batch * batch_size +
batch_size][:],theta = embedding.forward_db_visual(batch_data)
# emb,theta = backbone(batch.to(device),visualize=True)#.cpu()
# over_lap=calculate_overlap(theta,patch_size=16)
if overlap==True:
over_lap=calculate_overlap_near(theta,patch_size=28)
# pdb.set_trace()
over_lap_all+=over_lap
# embeddings[idx:idx + batch_size] = emb.cpu()
# pdb.set_trace()
# utils.save_patch(batch_data,batch_data,theta,patch_size=embedding.model.patch_size,save_folder=logpath,iter1=batch,epoch=0,step=0)
batch += 1
faceness_scores.append(name_lmk_score[-1])
batch_data = np.empty((2 * rare_size, 3, 112, 112))
embedding = Embedding(model_path, data_shape, rare_size)
for img_index, each_line in enumerate(files[len(files) - rare_size:]):
name_lmk_score = each_line.strip().split(' ')
img_name = os.path.join(img_path, name_lmk_score[0])
img = cv2.imread(img_name)
lmk = np.array([float(x) for x in name_lmk_score[1:-1]],
dtype=np.float32)
lmk = lmk.reshape((5, 2))
input_blob = embedding.get(img, lmk)
batch_data[2 * img_index][:] = input_blob[0]
batch_data[2 * img_index + 1][:] = input_blob[1]
if (img_index + 1) % rare_size == 0:
print('batch', batch)
if not save_samples:
img_feats[len(files) -
rare_size:][:] = embedding.forward_db(batch_data)
else:
# if (visualize==True) and (overlap==True):
img_feats[batch * batch_size:batch * batch_size +
batch_size][:],theta = embedding.forward_db_visual(batch_data)
# emb,theta = backbone(batch.to(device),visualize=True)#.cpu()
# over_lap=calculate_overlap(theta,patch_size=16)
if overlap==True:
over_lap=calculate_overlap_near(theta,patch_size=28)
# pdb.set_trace()
over_lap_all+=over_lap
# embeddings[idx:idx + batch_size] = emb.cpu()
# save_patch(batch_data,batch,theta,patch_size=embedding.backbone.patch_size,save_folder=logpath,iter1=batch_count,epoch=epoch,step=step)
batch += 1
faceness_scores.append(name_lmk_score[-1])
faceness_scores = np.array(faceness_scores).astype(np.float32)
# img_feats = np.ones( (len(files), 1024), dtype=np.float32) * 0.01
# faceness_scores = np.ones( (len(files), ), dtype=np.float32 )
over_lap_mean=np.mean(over_lap_all)
over_lap_var=np.var(over_lap_all)
print ('mean:'+str(over_lap_mean))
print ('var:'+str(over_lap_var))
return img_feats, faceness_scores
# In[ ]:
def image2template_feature(img_feats=None, templates=None, medias=None):
# ==========================================================
# 1. face image feature l2 normalization. img_feats:[number_image x feats_dim]
# 2. compute media feature.
# 3. compute template feature.
# ==========================================================
unique_templates = np.unique(templates)
template_feats = np.zeros((len(unique_templates), img_feats.shape[1]))
for count_template, uqt in enumerate(unique_templates):
(ind_t,) = np.where(templates == uqt)
face_norm_feats = img_feats[ind_t]
face_medias = medias[ind_t]
unique_medias, unique_media_counts = np.unique(face_medias,
return_counts=True)
media_norm_feats = []
for u, ct in zip(unique_medias, unique_media_counts):
(ind_m,) = np.where(face_medias == u)
if ct == 1:
media_norm_feats += [face_norm_feats[ind_m]]
else: # image features from the same video will be aggregated into one feature
media_norm_feats += [
np.mean(face_norm_feats[ind_m], axis=0, keepdims=True)
]
media_norm_feats = np.array(media_norm_feats)
# media_norm_feats = media_norm_feats / np.sqrt(np.sum(media_norm_feats ** 2, -1, keepdims=True))
template_feats[count_template] = np.sum(media_norm_feats, axis=0)
if count_template % 2000 == 0:
print('Finish Calculating {} template features.'.format(
count_template))
# template_norm_feats = template_feats / np.sqrt(np.sum(template_feats ** 2, -1, keepdims=True))
template_norm_feats = sklearn.preprocessing.normalize(template_feats)
# print(template_norm_feats.shape)
return template_norm_feats, unique_templates
# In[ ]:
def verification(template_norm_feats=None,
unique_templates=None,
p1=None,
p2=None):
# ==========================================================
# Compute set-to-set Similarity Score.
# ==========================================================
template2id = np.zeros((max(unique_templates) + 1, 1), dtype=int)
for count_template, uqt in enumerate(unique_templates):
template2id[uqt] = count_template
score = np.zeros((len(p1),)) # save cosine distance between pairs
total_pairs = np.array(range(len(p1)))
batchsize = 100000 # small batchsize instead of all pairs in one batch due to the memory limiation
sublists = [
total_pairs[i:i + batchsize] for i in range(0, len(p1), batchsize)
]
total_sublists = len(sublists)
for c, s in enumerate(sublists):
feat1 = template_norm_feats[template2id[p1[s]]]
feat2 = template_norm_feats[template2id[p2[s]]]
similarity_score = np.sum(feat1 * feat2, -1)
score[s] = similarity_score.flatten()
if c % 10 == 0:
print('Finish {}/{} pairs.'.format(c, total_sublists))
return score
# In[ ]:
def verification2(template_norm_feats=None,
unique_templates=None,
p1=None,
p2=None):
template2id = np.zeros((max(unique_templates) + 1, 1), dtype=int)
for count_template, uqt in enumerate(unique_templates):
template2id[uqt] = count_template
score = np.zeros((len(p1),)) # save cosine distance between pairs
total_pairs = np.array(range(len(p1)))
batchsize = 100000 # small batchsize instead of all pairs in one batch due to the memory limiation
sublists = [
total_pairs[i:i + batchsize] for i in range(0, len(p1), batchsize)
]
total_sublists = len(sublists)
for c, s in enumerate(sublists):
feat1 = template_norm_feats[template2id[p1[s]]]
feat2 = template_norm_feats[template2id[p2[s]]]
similarity_score = np.sum(feat1 * feat2, -1)
score[s] = similarity_score.flatten()
if c % 10 == 0:
print('Finish {}/{} pairs.'.format(c, total_sublists))
return score
def convert_onnx(net, path_module, output_path,output_name='model_196', opset=12, simplify=False):
output=os.path.join(output_path, "%s.onnx" % output_name)
assert isinstance(net, torch.nn.Module)
img = np.random.randint(0, 255, size=(112, 112, 3), dtype=np.int32)
img = img.astype(np.float)
img = (img / 255. - 0.5) / 0.5 # torch style norm
img = img.transpose((2, 0, 1))
img = torch.from_numpy(img).unsqueeze(0).float()
net=net.to('cpu')
# weight = torch.load(path_module)
# net.load_state_dict(weight)
# net.eval()
# torch.onnx.export(net, img, output, keep_initializers_as_inputs=False, verbose=False, opset_version=opset)
pdb.set_trace()
torch.onnx.export(net.module, img, output, keep_initializers_as_inputs=False, verbose=False, opset_version=opset)
model = onnx.load(output)
graph = model.graph
graph.input[0].type.tensor_type.shape.dim[0].dim_param = 'None'
if simplify:
from onnxsim import simplify
model, check = simplify(model)
assert check, "Simplified ONNX model could not be validated"
onnx.save(model, output)
def read_score(path):
with open(path, 'rb') as fid:
img_feats = pickle.load(fid)
return img_feats
def load_part_checkpoint_landmark(path,model,pretrain_name=['stn','output']):
# pdb.set_trace()
pretrained_dict = torch.load(path, map_location='cpu')
model_dict = model.state_dict()
# 1. filter out unnecessary keys
# pretrained_dict=list(pretrained_dict.keys())
back_remove=list(pretrained_dict.keys())
for keys in back_remove:
if 'dummy_orthogonal_classifier' in keys:
# pdb.set_trace()
continue
pretrained_dict[keys.replace('module.','')]=pretrained_dict.pop(keys)
# pdb.set_trace()
# for name_space in pretrain_name:
pretrained_dict = {k: v for k, v in pretrained_dict.items() if pretrain_name[0] in k or pretrain_name[1] in k}
# pretrained_dict = {k: v for k, v in pretrained_dict.items() if pretrain_name[0] in k or pretrain_name[1] in k}
# pretrained_dict = {k: v for k, v in pretrained_dict.items() if k in model_dict}
# 2. overwrite entries in the existing state dict
model_dict.update(pretrained_dict)
# 3. load the new state dict
model.load_state_dict(model_dict,strict=True)
# model.encoder.output_layer.load_state_dict(pretrained_dict,strict=True)
model_dict = model.state_dict()
#freeze stn and output layer
for name, param in model.named_parameters():
# if not param.requires_grad:
if pretrain_name[0] in name or pretrain_name[1] in name:
# pdb.set_trace()
param.requires_grad = False
# # Step1: Load Meta Data
# In[ ]:
assert target == 'IJBC' or target == 'IJBB'
# =============================================================
# load image and template relationships for template feature embedding
# tid --> template id, mid --> media id
# format:
# image_name tid mid
# =============================================================
start = timeit.default_timer()
templates, medias = read_template_media_list(
os.path.join('%s/meta' % image_path,
'%s_face_tid_mid.txt' % target.lower()))
stop = timeit.default_timer()
print('Time: %.2f s. ' % (stop - start))
# In[ ]:
# =============================================================
# load template pairs for template-to-template verification
# tid : template id, label : 1/0
# format:
# tid_1 tid_2 label
# =============================================================
start = timeit.default_timer()
p1, p2, label = read_template_pair_list(
os.path.join('%s/meta' % image_path,
'%s_template_pair_label.txt' % target.lower()))
stop = timeit.default_timer()
print('Time: %.2f s. ' % (stop - start))
# # Step 2: Get Image Features
# In[ ]:
# =============================================================
# load image features
# format:
# img_feats: [image_num x feats_dim] (227630, 512)
# =============================================================
start = timeit.default_timer()
img_path = '%s/loose_crop' % image_path
img_list_path = '%s/meta/%s_name_5pts_score.txt' % (image_path, target.lower())
img_list = open(img_list_path)
files = img_list.readlines()
# files_list = divideIntoNstrand(files, rank_size)
files_list = files
# img_feats
# for i in range(rank_size):
img_feats, faceness_scores = get_image_feature(img_path, files_list,
model_path, 0, gpu_id)
stop = timeit.default_timer()
print('Time: %.2f s. ' % (stop - start))
print('Feature Shape: ({} , {}) .'.format(img_feats.shape[0],
img_feats.shape[1]))
# # Step3: Get Template Features
# In[ ]:
# =============================================================
# compute template features from image features.
# =============================================================
start = timeit.default_timer()
# ==========================================================
# Norm feature before aggregation into template feature?
# Feature norm from embedding network and faceness score are able to decrease weights for noise samples (not face).
# ==========================================================
# 1. FaceScore (Feature Norm)
# 2. FaceScore (Detector)
if use_flip_test:
# concat --- F1
# img_input_feats = img_feats
# add --- F2
img_input_feats = img_feats[:, 0:img_feats.shape[1] //
2] + img_feats[:, img_feats.shape[1] // 2:]
else:
img_input_feats = img_feats[:, 0:img_feats.shape[1] // 2]
if use_norm_score:
img_input_feats = img_input_feats
else:
# normalise features to remove norm information
img_input_feats = img_input_feats / np.sqrt(
np.sum(img_input_feats ** 2, -1, keepdims=True))
if use_detector_score:
print(img_input_feats.shape, faceness_scores.shape)
img_input_feats = img_input_feats * faceness_scores[:, np.newaxis]
else:
img_input_feats = img_input_feats
template_norm_feats, unique_templates = image2template_feature(
img_input_feats, templates, medias)
stop = timeit.default_timer()
print('Time: %.2f s. ' % (stop - start))
# # Step 4: Get Template Similarity Scores
# In[ ]:
# =============================================================
# compute verification scores between template pairs.
# =============================================================
start = timeit.default_timer()
score = verification(template_norm_feats, unique_templates, p1, p2)
stop = timeit.default_timer()
print('Time: %.2f s. ' % (stop - start))
# In[ ]:
save_path = os.path.join(result_dir, args.job)
# save_path = result_dir + '/%s_result' % target
if not os.path.exists(save_path):
os.makedirs(save_path)
score_save_file = os.path.join(save_path, "%s.npy" % target.lower())
np.save(score_save_file, score)
# # Step 5: Get ROC Curves and TPR@FPR Table
# In[ ]:
files = [score_save_file]
methods = []
scores = []
for file in files:
methods.append(Path(file).stem)
scores.append(np.load(file))
methods = np.array(methods)
scores = dict(zip(methods, scores))
colours = dict(
zip(methods, sample_colours_from_colourmap(methods.shape[0], 'Set2')))
x_labels = [10 ** -6, 10 ** -5, 10 ** -4, 10 ** -3, 10 ** -2, 10 ** -1]
tpr_fpr_table = PrettyTable(['Methods'] + [str(x) for x in x_labels])
fig = plt.figure()
for method in methods:
fpr, tpr, _ = roc_curve(label, scores[method])
roc_auc = auc(fpr, tpr)
fpr = np.flipud(fpr)
tpr = np.flipud(tpr) # select largest tpr at same fpr
plt.plot(fpr,
tpr,
color=colours[method],
lw=1,
label=('[%s (AUC = %0.4f %%)]' %
(method.split('-')[-1], roc_auc * 100)))
tpr_fpr_row = []
tpr_fpr_row.append("%s-%s" % (method, target))
for fpr_iter in np.arange(len(x_labels)):
_, min_index = min(
list(zip(abs(fpr - x_labels[fpr_iter]), range(len(fpr)))))
tpr_fpr_row.append('%.2f' % (tpr[min_index] * 100))
tpr_fpr_table.add_row(tpr_fpr_row)
plt.xlim([10 ** -6, 0.1])
plt.ylim([0.3, 1.0])
plt.grid(linestyle='--', linewidth=1)
plt.xticks(x_labels)
plt.yticks(np.linspace(0.3, 1.0, 8, endpoint=True))
plt.xscale('log')
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('ROC on IJB')
plt.legend(loc="lower right")
fig.savefig(os.path.join(save_path, '%s.pdf' % target.lower()))
print(tpr_fpr_table)
print(model_path)