-
Notifications
You must be signed in to change notification settings - Fork 51
/
MSDN.py
641 lines (524 loc) · 28.8 KB
/
MSDN.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
import cv2
import numpy as np
import numpy.random as npr
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.autograd import Variable
import torch.utils.model_zoo as model_zoo
import torchvision.models as models
import os.path as osp
from utils.timer import Timer
from utils.HDN_utils import check_relationship_recall
from fast_rcnn.nms_wrapper import nms
from rpn_msr.proposal_layer import proposal_layer as proposal_layer_py
from rpn_msr.anchor_target_layer import anchor_target_layer as anchor_target_layer_py
from rpn_msr.proposal_target_layer_hdn import proposal_target_layer as proposal_target_layer_py
from fast_rcnn.bbox_transform import bbox_transform_inv_hdn, clip_boxes
from fast_rcnn.hierarchical_message_passing_structure import Hierarchical_Message_Passing_Structure
from Language_Model import Language_Model
from RPN import RPN
from fast_rcnn.config import cfg
from utils.cython_bbox import bbox_overlaps
import network
from network import Conv2d, FC
# from roi_pooling.modules.roi_pool_py import RoIPool
from roi_pooling.modules.roi_pool import RoIPool
from vgg16 import VGG16
from MSDN_base import HDN_base
import pdb
DEBUG = False
TIME_IT = cfg.TIME_IT
def nms_detections(pred_boxes, scores, nms_thresh, inds=None):
dets = np.hstack((pred_boxes,
scores[:, np.newaxis])).astype(np.float32)
keep = nms(dets, nms_thresh)
if inds is None:
return pred_boxes[keep], scores[keep], keep
return pred_boxes[keep], scores[keep], inds[keep], keep
class Hierarchical_Descriptive_Model(HDN_base):
def __init__(self,nhidden, n_object_cats, n_predicate_cats, n_vocab, voc_sign,
max_word_length, MPS_iter, use_language_loss, object_loss_weight,
predicate_loss_weight,
dropout=False,
use_kmeans_anchors=False,
gate_width=128,
nhidden_caption=256,
nembedding = 256,
rnn_type='LSTM_normal',
rnn_droptout=0.0, rnn_bias=False,
use_region_reg=False, use_kernel=False):
super(Hierarchical_Descriptive_Model, self).__init__(nhidden, n_object_cats, n_predicate_cats, n_vocab, voc_sign,
max_word_length, MPS_iter, use_language_loss, object_loss_weight, predicate_loss_weight,
dropout, use_kmeans_anchors, nhidden_caption, nembedding, rnn_type, use_region_reg)
self.rpn = RPN(use_kmeans_anchors)
self.roi_pool_object = RoIPool(7, 7, 1.0/16)
self.roi_pool_phrase = RoIPool(7, 7, 1.0/16)
self.roi_pool_region = RoIPool(7, 7, 1.0/16)
self.fc6_obj = FC(512 * 7 * 7, nhidden, relu=True)
self.fc7_obj = FC(nhidden, nhidden, relu=False)
self.fc6_phrase = FC(512 * 7 * 7, nhidden, relu=True)
self.fc7_phrase = FC(nhidden, nhidden, relu=False)
self.fc6_region = FC(512 * 7 * 7, nhidden, relu=True)
self.fc7_region = FC(nhidden, nhidden, relu=False)
if MPS_iter == 0:
self.mps = None
else:
self.mps = Hierarchical_Message_Passing_Structure(nhidden, dropout,
gate_width=gate_width, use_kernel_function=use_kernel) # the hierarchical message passing structure
network.weights_normal_init(self.mps, 0.01)
self.score_obj = FC(nhidden, self.n_classes_obj, relu=False)
self.bbox_obj = FC(nhidden, self.n_classes_obj * 4, relu=False)
self.score_pred = FC(nhidden, self.n_classes_pred, relu=False)
if self.use_region_reg:
self.bbox_region = FC(nhidden, 4, relu=False)
network.weights_normal_init(self.bbox_region, 0.01)
else:
self.bbox_region = None
self.objectiveness = FC(nhidden, 2, relu=False)
if use_language_loss:
self.caption_prediction = \
Language_Model(rnn_type=self.rnn_type, ntoken=self.n_vocab, nimg=self.nhidden, nhidden=self.nhidden_caption,
nembed=self.nembedding, nlayers=2, nseq=self.max_word_length, voc_sign = self.voc_sign,
bias=rnn_bias, dropout=rnn_droptout)
else:
self.caption_prediction = Language_Model(rnn_type=self.rnn_type, ntoken=self.n_vocab, nimg=1, nhidden=1,
nembed=1, nlayers=1, nseq=1, voc_sign = self.voc_sign) # just to make the program run
network.weights_normal_init(self.score_obj, 0.01)
network.weights_normal_init(self.bbox_obj, 0.005)
network.weights_normal_init(self.score_pred, 0.01)
network.weights_normal_init(self.objectiveness, 0.01)
self.objectiveness_loss = None
def forward(self, im_data, im_info, gt_objects=None, gt_relationships=None, gt_regions=None,
use_beam_search=False, graph_generation=False):
self.timer.tic()
features, object_rois, region_rois = self.rpn(im_data, im_info, gt_objects, gt_regions)
if not self.training and gt_objects is not None:
zeros = np.zeros((gt_objects.shape[0], 1), dtype=gt_objects.dtype)
object_rois_gt = np.hstack((zeros, gt_objects[:, :4]))
object_rois_gt = network.np_to_variable(object_rois_gt, is_cuda=True)
object_rois[:object_rois_gt.size(0)] = object_rois_gt
if not self.training and gt_regions is not None:
zeros = np.zeros((gt_regions.shape[0], 1), dtype=gt_regions.dtype)
region_rois = np.hstack((zeros, gt_regions[:, :4]))
region_rois = network.np_to_variable(region_rois, is_cuda=True)
# print 'region_rois[gt]:', region_rois
# print 'object_rois.shape', object_rois.size()
# print 'features.std'
# print features.data.std()
if TIME_IT:
torch.cuda.synchronize()
print '\t[RPN]: %.3fs' % self.timer.toc(average=False)
self.timer.tic()
roi_data_object, roi_data_predicate, roi_data_region, mat_object, mat_phrase, mat_region = \
self.proposal_target_layer(object_rois, region_rois, gt_objects, gt_relationships, gt_regions,
self.n_classes_obj, self.voc_sign, self.training, graph_generation=graph_generation)
if TIME_IT:
torch.cuda.synchronize()
print '\t[Proposal]: %.3fs' % self.timer.toc(average=False)
self.timer.tic()
object_rois = roi_data_object[0]
phrase_rois = roi_data_predicate[0]
region_rois = roi_data_region[0]
# print 'object_rois_num: {}'.format(object_rois.size()[0])
# print 'phrase_rois_num: {}'.format(phrase_rois.size()[0])
# print 'region_rois_num: {}'.format(region_rois.size()[0])
# roi pool
pooled_object_features = self.roi_pool_object(features, object_rois)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[object_pooling]: %.3fs' % self.timer.toc(average=False)
#print 'pool5_object.std'
#print pooled_object_features.data.std()
pooled_object_features = pooled_object_features.view(pooled_object_features.size()[0], -1)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[object_feature_view]: %.3fs' % self.timer.toc(average=False)
pooled_object_features = self.fc6_obj(pooled_object_features)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[object_feature_fc6]: %.3fs' % self.timer.toc(average=False)
if self.dropout:
pooled_object_features = F.dropout(pooled_object_features, training = self.training)
#print 'fc6_object.std'
#print pooled_object_features.data.std()
pooled_object_features = self.fc7_obj(pooled_object_features)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[object_feature_fc7]: %.3fs' % self.timer.toc(average=False)
if self.dropout:
pooled_object_features = F.dropout(pooled_object_features, training = self.training)
#print 'fc7_object.std'
#print pooled_object_features.data.std()
pooled_phrase_features = self.roi_pool_phrase(features, phrase_rois)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[phrase_pooling]: %.3fs' % self.timer.toc(average=False)
#print 'pool5_phrase.std'
#print pooled_phrase_features.data.std()
pooled_phrase_features = pooled_phrase_features.view(pooled_phrase_features.size()[0], -1)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[phrase_feature_view]: %.3fs' % self.timer.toc(average=False)
pooled_phrase_features = self.fc6_phrase(pooled_phrase_features)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[phrase_feature_fc6]: %.3fs' % self.timer.toc(average=False)
if self.dropout:
pooled_phrase_features = F.dropout(pooled_phrase_features, training = self.training)
#print 'fc6_phrase.std'
#print pooled_phrase_features.data.std()
pooled_phrase_features = self.fc7_phrase(pooled_phrase_features)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[phrase_feature_fc7]: %.3fs' % self.timer.toc(average=False)
if self.dropout:
pooled_phrase_features = F.dropout(pooled_phrase_features, training = self.training)
#print 'fc7_phrase.std'
#print pooled_phrase_features.data.std()
pooled_region_features = self.roi_pool_region(features, region_rois)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[region_pooling]: %.3fs' % self.timer.toc(average=False)
#print 'pool5_region.std'
#print pooled_region_features.data.std()
pooled_region_features = pooled_region_features.view(pooled_region_features.size()[0], -1)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[region_feature_view]: %.3fs' % self.timer.toc(average=False)
pooled_region_features = self.fc6_region(pooled_region_features)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[region_feature_fc6]: %.3fs' % self.timer.toc(average=False)
if self.dropout:
pooled_region_features = F.dropout(pooled_region_features, training = self.training)
#print 'fc6_region.std'
#print pooled_region_features.data.std()
pooled_region_features = self.fc7_region(pooled_region_features)
if TIME_IT:
torch.cuda.synchronize()
print '\t\t[region_feature_fc7]: %.3fs' % self.timer.toc(average=False)
if self.dropout:
pooled_region_features = F.dropout(pooled_region_features, training = self.training)
#print 'fc7_region.std'
#print pooled_region_features.data.std()
# print 'pre_mps_object.std', pooled_object_features.data.std()
# print 'pre_mps_phrase.std', pooled_phrase_features.data.std()
# print 'pre_mps_region.std', pooled_region_features.data.std()
# bounding box regression before message passing
bbox_object = self.bbox_obj(F.relu(pooled_object_features))
if self.use_region_reg:
bbox_region = self.bbox_region(F.relu(pooled_region_features))
if TIME_IT:
torch.cuda.synchronize()
print '\t[Pre-MPS]: %.3fs' % self.timer.toc(average=False)
self.timer.tic()
# hierarchical message passing structure
if self.MPS_iter < 0:
if self.training:
self.MPS_iter = npr.choice(self.MPS_iter_range)
else:
self.MPS_iter = cfg.TEST.MPS_ITER_NUM
for i in xrange(self.MPS_iter):
pooled_object_features, pooled_phrase_features, pooled_region_features = \
self.mps(pooled_object_features, pooled_phrase_features, pooled_region_features, \
mat_object, mat_phrase, mat_region)
if TIME_IT:
torch.cuda.synchronize()
print '\t[Passing]: %.3fs' % self.timer.toc(average=False)
# print 'post_mps_object.std', pooled_object_features.data.std()
# print 'post_mps_phrase.std', pooled_phrase_features.data.std()
# print 'post_mps_region.std', pooled_region_features.data.std()
self.timer.tic()
pooled_object_features = F.relu(pooled_object_features)
pooled_phrase_features = F.relu(pooled_phrase_features)
pooled_region_features = F.relu(pooled_region_features)
cls_score_object = self.score_obj(pooled_object_features)
cls_prob_object = F.softmax(cls_score_object)
cls_score_predicate = self.score_pred(pooled_phrase_features)
cls_prob_predicate = F.softmax(cls_score_predicate)
if not self.use_region_reg:
bbox_region = Variable(torch.zeros(pooled_region_features.size(0), 4).cuda())
cls_objectiveness_region = self.objectiveness(pooled_region_features)
# print 'cls_score_object.std', cls_score_object.data.std()
# print 'cls_pred_box.std', bbox_object.data.std()
# print 'cls_score_phrase.std', cls_score_predicate.data.std()
if TIME_IT:
torch.cuda.synchronize()
print '\t[Post-MPS]: %.3fs' % self.timer.toc(average=False)
# if DEBUG:
# print 'cls_score_predicate'
# print cls_score_predicate
# print 'roi_data_predicate[1]'
# print roi_data_predicate[1]
if self.training:
self.cross_entropy_object, self.loss_obj_box = self.build_loss_object(cls_score_object, bbox_object, roi_data_object)
self.cross_entropy_predicate, self.tp_pred, self.tf_pred, self.fg_cnt_pred, self.bg_cnt_pred = \
self.build_loss_cls(cls_score_predicate, roi_data_predicate[1])
# print 'accuracy: %2.2f%%' % (((self.tp_pred + self.tf_pred) / float(self.fg_cnt_pred + self.bg_cnt_pred)) * 100)
# self.timer.tic()
if self.use_language_loss:
self.region_caption_loss = self.caption_prediction(pooled_region_features, roi_data_region[1])
else:
self.region_caption_loss = Variable(torch.zeros(1).cuda())
if self.use_region_reg:
self.loss_region_box = self.build_loss_bbox(bbox_region, roi_data_region)
# print '\t[Caption]: %.3fs' % self.timer.toc(average=False)
region_caption = None
self.objectiveness_loss = self.build_loss_objectiveness(cls_objectiveness_region, \
roi_data_region[3][:, 0].ne(0).type(torch.cuda.LongTensor))
else:
# assert False, 'Have not implemented!\n'
if self.use_language_loss:
# region_caption, caption_logprobs = self.caption_prediction.beamsearch(pooled_region_features, 10)
if use_beam_search:
search_func = self.caption_prediction.beamsearch
else:
search_func = self.caption_prediction.baseline_search
region_caption = search_func(pooled_region_features, 5)
# pdb.set_trace()
else:
region_caption = None
caption_logprobs = None
caption_logprobs = F.log_softmax(cls_objectiveness_region)[:, 1].squeeze().cpu().data
return (cls_prob_object, bbox_object, object_rois), \
(cls_prob_predicate, mat_phrase), \
(region_caption, bbox_region, region_rois, caption_logprobs)
@staticmethod
def proposal_target_layer(object_rois, region_rois, gt_objects, gt_relationships,
gt_regions, n_classes_obj, voc_sign, is_training=False, graph_generation=False):
"""
----------
object_rois: (1 x H x W x A, 5) [0, x1, y1, x2, y2]
region_rois: (1 x H x W x A, 5) [0, x1, y1, x2, y2]
gt_objects: (G_obj, 5) [x1 ,y1 ,x2, y2, obj_class] int
gt_relationships: (G_obj, G_obj) [pred_class] int (-1 for no relationship)
gt_regions: (G_region, 4+40) [x1, y1, x2, y2, word_index] (-1 for padding)
# gt_ishard: (G_region, 4+40) {0 | 1} 1 indicates hard
# dontcare_areas: (D, 4) [ x1, y1, x2, y2]
n_classes_obj
n_classes_pred
is_training to indicate whether in training scheme
----------
Returns
----------
rois: (1 x H x W x A, 5) [0, x1, y1, x2, y2]
labels: (1 x H x W x A, 1) {0,1,...,_num_classes-1}
bbox_targets: (1 x H x W x A, K x4) [dx1, dy1, dx2, dy2]
bbox_inside_weights: (1 x H x W x A, Kx4) 0, 1 masks for the computing loss
bbox_outside_weights: (1 x H x W x A, Kx4) 0, 1 masks for the computing loss
"""
object_rois = object_rois.data.cpu().numpy()
region_rois = region_rois.data.cpu().numpy()
object_labels, object_rois, bbox_targets, bbox_inside_weights, bbox_outside_weights, mat_object, \
phrase_label, phrase_rois, mat_phrase, region_seq, region_rois, \
bbox_targets_region, bbox_inside_weights_region, bbox_outside_weights_region, mat_region= \
proposal_target_layer_py(object_rois, region_rois, gt_objects, gt_relationships,
gt_regions, n_classes_obj, voc_sign, is_training, graph_generation=graph_generation)
# print labels.shape, bbox_targets.shape, bbox_inside_weights.shape
if is_training:
object_labels = network.np_to_variable(object_labels, is_cuda=True, dtype=torch.LongTensor)
bbox_targets = network.np_to_variable(bbox_targets, is_cuda=True)
bbox_inside_weights = network.np_to_variable(bbox_inside_weights, is_cuda=True)
bbox_outside_weights = network.np_to_variable(bbox_outside_weights, is_cuda=True)
phrase_label = network.np_to_variable(phrase_label, is_cuda=True, dtype=torch.LongTensor)
region_seq = network.np_to_variable(region_seq, is_cuda=True, dtype=torch.LongTensor)
bbox_targets_region = network.np_to_variable(bbox_targets_region, is_cuda=True)
bbox_inside_weights_region = network.np_to_variable(bbox_inside_weights_region, is_cuda=True)
bbox_outside_weights_region = network.np_to_variable(bbox_outside_weights_region, is_cuda=True)
object_rois = network.np_to_variable(object_rois, is_cuda=True)
phrase_rois = network.np_to_variable(phrase_rois, is_cuda=True)
region_rois = network.np_to_variable(region_rois, is_cuda=True)
return (object_rois, object_labels, bbox_targets, bbox_inside_weights, bbox_outside_weights), \
(phrase_rois, phrase_label), \
(region_rois, region_seq, bbox_targets_region, bbox_inside_weights_region, bbox_outside_weights_region), \
mat_object, mat_phrase, mat_region
def interpret_HDN(self, cls_prob, bbox_pred, rois, cls_prob_predicate,
mat_phrase, im_info, nms=True, clip=True, min_score=0.0,
top_N=100, use_gt_boxes=False):
scores, inds = cls_prob[:, 1:].data.max(1)
inds += 1
scores, inds = scores.cpu().numpy(), inds.cpu().numpy()
predicate_scores, predicate_inds = cls_prob_predicate[:, 1:].data.max(1)
predicate_inds += 1
predicate_scores, predicate_inds = predicate_scores.cpu().numpy(), predicate_inds.cpu().numpy()
keep = np.where((inds > 0) & (scores >= min_score))
scores, inds = scores[keep], inds[keep]
# Apply bounding-box regression deltas
keep = keep[0]
box_deltas = bbox_pred.data.cpu().numpy()[keep]
box_deltas = np.asarray([
box_deltas[i, (inds[i] * 4): (inds[i] * 4 + 4)] for i in range(len(inds))
], dtype=np.float)
boxes = rois.data.cpu().numpy()[keep, 1:5] / im_info[0][2]
if use_gt_boxes:
nms = False
clip = False
pred_boxes = boxes
else:
pred_boxes = bbox_transform_inv_hdn(boxes, box_deltas)
if clip:
pred_boxes = clip_boxes(pred_boxes, im_info[0][:2] / im_info[0][2])
# nms
if nms and pred_boxes.shape[0] > 0:
pred_boxes, scores, inds, keep_keep = nms_detections(pred_boxes, scores, 0.60, inds=inds)
keep = keep[keep_keep]
sub_list = np.array([], dtype=int)
obj_list = np.array([], dtype=int)
pred_list = np.array([], dtype=int)
# print 'keep', keep
# print 'mat_phrase', mat_phrase
for i in range(mat_phrase.shape[0]):
sub_id = np.where(keep == mat_phrase[i, 0])[0]
obj_id = np.where(keep == mat_phrase[i, 1])[0]
if len(sub_id) > 0 and len(obj_id) > 0:
sub_list = np.append(sub_list, sub_id[0])
obj_list = np.append(obj_list, obj_id[0])
pred_list = np.append(pred_list, i)
total_scores = predicate_scores.squeeze()[pred_list] \
* scores[sub_list].squeeze() * scores[obj_list].squeeze()
top_N_list = total_scores.argsort()[::-1][:top_N]
predicate_inds = predicate_inds.squeeze()[pred_list[top_N_list]]
subject_inds = inds[sub_list[top_N_list]]
object_inds = inds[obj_list[top_N_list]]
subject_boxes = pred_boxes[sub_list[top_N_list]]
object_boxes = pred_boxes[obj_list[top_N_list]]
return pred_boxes, scores, inds, subject_inds, object_inds, subject_boxes, object_boxes, predicate_inds
def interpret_result(self, cls_prob, bbox_pred, rois, cls_prob_predicate,
mat_phrase, im_info, im_shape, nms=True, clip=True, min_score=0.01,
use_gt_boxes=False):
scores, inds = cls_prob[:, 0:].data.max(1)
# inds += 1
scores, inds = scores.cpu().numpy(), inds.cpu().numpy()
predicate_scores, predicate_inds = cls_prob_predicate[:, 0:].data.max(1)
# predicate_inds += 1
predicate_scores, predicate_inds = predicate_scores.cpu().numpy(), predicate_inds.cpu().numpy()
keep = np.where((inds > 0) & (scores >= min_score))
scores, inds = scores[keep], inds[keep]
# Apply bounding-box regression deltas
keep = keep[0]
box_deltas = bbox_pred.data.cpu().numpy()[keep]
box_deltas = np.asarray([
box_deltas[i, (inds[i] * 4): (inds[i] * 4 + 4)] for i in range(len(inds))
], dtype=np.float)
boxes = rois.data.cpu().numpy()[keep, 1:5] / im_info[0][2]
if use_gt_boxes:
nms = False
clip = False
pred_boxes = boxes
else:
pred_boxes = bbox_transform_inv_hdn(boxes, box_deltas)
if clip:
pred_boxes = clip_boxes(pred_boxes, im_shape)
# nms
if nms and pred_boxes.shape[0] > 0:
pred_boxes, scores, inds, keep_keep = nms_detections(pred_boxes, scores, 0.3, inds=inds)
keep = keep[keep_keep]
sub_list = np.array([], dtype=int)
obj_list = np.array([], dtype=int)
pred_list = np.array([], dtype=int)
# print 'keep', keep
# print 'mat_phrase', mat_phrase
for i in range(mat_phrase.shape[0]):
sub_id = np.where(keep == mat_phrase[i, 0])[0]
obj_id = np.where(keep == mat_phrase[i, 1])[0]
if len(sub_id) > 0 and len(obj_id) > 0:
sub_list = np.append(sub_list, sub_id[0])
obj_list = np.append(obj_list, obj_id[0])
pred_list = np.append(pred_list, i)
predicate_scores = predicate_scores.squeeze()[pred_list]
final_list = predicate_scores.argsort()[::-1]
predicate_inds = predicate_inds.squeeze()[pred_list[final_list]]
sub_list = sub_list[final_list]
obj_list = obj_list[final_list]
region_list = mat_phrase[pred_list[final_list], 2:]
return pred_boxes, scores, inds, sub_list, obj_list, predicate_inds, region_list
def caption(self, im_path, gt_objects=None, gt_regions=None, thr=0.0, nms=False, top_N=100, clip=True, use_beam_search=False):
image = cv2.imread(im_path)
# print 'image.shape', image.shape
im_data, im_scales = self.get_image_blob_noscale(image)
# print 'im_data.shape', im_data.shape
# print 'im_scales', im_scales
if gt_objects is not None:
gt_objects[:, :4] = gt_objects[:, :4] * im_scales[0]
if gt_regions is not None:
gt_regions[:, :4] = gt_regions[:, :4] * im_scales[0]
im_info = np.array(
[[im_data.shape[1], im_data.shape[2], im_scales[0]]],
dtype=np.float32)
# pdb.set_trace()
region_result = self(im_data, im_info, gt_objects, gt_regions=gt_regions, use_beam_search=use_beam_search)[2]
region_caption, bbox_pred, region_rois, logprobs = region_result[:]
boxes = region_rois.data.cpu().numpy()[:, 1:5] / im_info[0][2]
box_deltas = bbox_pred.data.cpu().numpy()
pred_boxes = bbox_transform_inv_hdn(boxes, box_deltas)
if clip:
pred_boxes = clip_boxes(pred_boxes, image.shape)
# print 'im_scales[0]', im_scales[0]
return (region_caption.numpy(), logprobs.numpy(), pred_boxes)
def describe(self, im_path, top_N=10):
image = cv2.imread(im_path)
# print 'image.shape', image.shape
im_data, im_scales = self.get_image_blob_noscale(image)
# print 'im_data.shape', im_data.shape
# print 'im_scales', im_scales
im_info = np.array(
[[im_data.shape[1], im_data.shape[2], im_scales[0]]],
dtype=np.float32)
object_result, predicate_result, region_result = self(im_data, im_info)
object_boxes, object_scores, object_inds, sub_assignment, obj_assignment, predicate_inds, region_assignment\
= self.interpret_result(object_result[0], object_result[1], object_result[2], \
predicate_result[0], predicate_result[1], \
im_info, image.shape)
region_caption, bbox_pred, region_rois, logprobs = region_result[:]
boxes = region_rois.data.cpu().numpy()[:, 1:5] / im_info[0][2]
box_deltas = bbox_pred.data.cpu().numpy()
pred_boxes = bbox_transform_inv_hdn(boxes, box_deltas)
pred_boxes = clip_boxes(pred_boxes, image.shape)
# print 'im_scales[0]', im_scales[0]
return (region_caption.numpy(), logprobs.numpy(), pred_boxes, \
object_boxes, object_inds, object_scores, \
sub_assignment, obj_assignment, predicate_inds, region_assignment)
def evaluate(self, im_data, im_info, gt_objects, gt_relationships, gt_regions,
thr=0.5, nms=False, top_Ns = [100], use_gt_boxes=False, use_gt_regions=False, only_predicate=False):
if use_gt_boxes:
gt_boxes_object = gt_objects[:, :4] * im_info[2]
else:
gt_boxes_object = None
if use_gt_regions:
gt_boxes_regions = gt_regions[:, :4] * im_info[2]
else:
gt_boxes_regions = None
object_result, predicate_result, region_result = \
self(im_data, im_info, gt_boxes_object, gt_regions=gt_boxes_regions, graph_generation=True)
cls_prob_object, bbox_object, object_rois = object_result[:3]
cls_prob_predicate, mat_phrase = predicate_result[:2]
# interpret the model output
obj_boxes, obj_scores, obj_inds, subject_inds, object_inds, \
subject_boxes, object_boxes, predicate_inds = \
self.interpret_HDN(cls_prob_object, bbox_object, object_rois,
cls_prob_predicate, mat_phrase, im_info,
nms=nms, top_N=max(top_Ns), use_gt_boxes=use_gt_boxes)
gt_objects[:, :4] /= im_info[0][2]
rel_cnt, rel_correct_cnt = check_relationship_recall(gt_objects, gt_relationships,
subject_inds, object_inds, predicate_inds,
subject_boxes, object_boxes, top_Ns, thres=thr,
only_predicate=only_predicate)
return rel_cnt, rel_correct_cnt
def build_loss_objectiveness(self, region_objectiveness, targets):
loss_objectiveness = F.cross_entropy(region_objectiveness, targets)
maxv, predict = region_objectiveness.data.max(1)
labels = targets.squeeze()
fg_cnt = torch.sum(labels.data.ne(0))
bg_cnt = labels.data.numel() - fg_cnt
if fg_cnt > 0:
self.tp_reg = torch.sum(predict[:fg_cnt].eq(labels.data[:fg_cnt]))
else:
self.tp_reg = 0.
if bg_cnt > 0:
self.tf_reg = torch.sum(predict[fg_cnt:].eq(labels.data[fg_cnt:]))
else:
self.tp_reg = 0.
self.fg_cnt_reg = fg_cnt
self.bg_cnt_reg = bg_cnt
return loss_objectiveness