-
Notifications
You must be signed in to change notification settings - Fork 207
/
cocostuff.py
793 lines (609 loc) · 26.1 KB
/
cocostuff.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
# based on
# https://github.com/kazuto1011/deeplab-pytorch/blob/master/libs/datasets
# /cocostuff.py
from __future__ import print_function
import os.path as osp
import pickle
from glob import glob
import cv2
import numpy as np
import scipy.io as sio
import torch
import torchvision.transforms as tvt
from PIL import Image
from torch.utils import data
from util import cocostuff_fine_to_coarse
from .util.cocostuff_fine_to_coarse import generate_fine_to_coarse
from ...utils.segmentation.render import render
from ...utils.segmentation.transforms import \
pad_and_or_crop, random_affine, custom_greyscale_numpy
__all__ = ["Coco10kFull", "Coco10kFew", "Coco164kFull", "Coco164kFew",
"Coco164kCuratedFew", "Coco164kCuratedFull"]
RENDER_DATA = False
class _Coco(data.Dataset):
"""Base class
This contains fields and methods common to all COCO datasets:
(COCO-fine) (182)
COCO-coarse (27)
COCO-few (6)
(COCOStuff-fine) (91)
COCOStuff-coarse (15)
COCOStuff-few (3)
For both 10k and 164k (though latter is unimplemented)
"""
def __init__(self, config=None, split=None, purpose=None, preload=False):
super(_Coco, self).__init__()
self.split = split
self.purpose = purpose
self.root = config.dataset_root
self.single_mode = hasattr(config, "single_mode") and config.single_mode
# always used (labels fields used to make relevancy mask for train)
self.gt_k = config.gt_k
self.pre_scale_all = config.pre_scale_all
self.pre_scale_factor = config.pre_scale_factor
self.input_sz = config.input_sz
self.include_rgb = config.include_rgb
self.no_sobel = config.no_sobel
assert ((not hasattr(config, "mask_input")) or (not config.mask_input))
self.mask_input = False
# only used if purpose is train
if purpose == "train":
self.use_random_scale = config.use_random_scale
if self.use_random_scale:
self.scale_max = config.scale_max
self.scale_min = config.scale_min
self.jitter_tf = tvt.ColorJitter(brightness=config.jitter_brightness,
contrast=config.jitter_contrast,
saturation=config.jitter_saturation,
hue=config.jitter_hue)
self.flip_p = config.flip_p # 0.5
self.use_random_affine = config.use_random_affine
if self.use_random_affine:
self.aff_min_rot = config.aff_min_rot
self.aff_max_rot = config.aff_max_rot
self.aff_min_shear = config.aff_min_shear
self.aff_max_shear = config.aff_max_shear
self.aff_min_scale = config.aff_min_scale
self.aff_max_scale = config.aff_max_scale
assert (not preload)
self.files = []
self.images = []
self.labels = []
if not osp.exists(config.fine_to_coarse_dict):
generate_fine_to_coarse(config.fine_to_coarse_dict)
with open(config.fine_to_coarse_dict, "rb") as dict_f:
d = pickle.load(dict_f)
self._fine_to_coarse_dict = d["fine_index_to_coarse_index"]
cv2.setNumThreads(0)
def _prepare_train(self, index, img, label):
# This returns gpu tensors.
# label is passed in canonical [0 ... 181] indexing
assert (img.shape[:2] == label.shape)
img = img.astype(np.float32)
label = label.astype(np.int32)
# shrink original images, for memory purposes, otherwise no point
if self.pre_scale_all:
assert (self.pre_scale_factor < 1.)
img = cv2.resize(img, dsize=None, fx=self.pre_scale_factor,
fy=self.pre_scale_factor,
interpolation=cv2.INTER_LINEAR)
label = cv2.resize(label, dsize=None, fx=self.pre_scale_factor,
fy=self.pre_scale_factor,
interpolation=cv2.INTER_NEAREST)
# basic augmentation transforms for both img1 and img2
if self.use_random_scale:
# bilinear interp requires float img
scale_factor = (np.random.rand() * (self.scale_max - self.scale_min)) + \
self.scale_min
img = cv2.resize(img, dsize=None, fx=scale_factor, fy=scale_factor,
interpolation=cv2.INTER_LINEAR)
label = cv2.resize(label, dsize=None, fx=scale_factor, fy=scale_factor,
interpolation=cv2.INTER_NEAREST)
# random crop to input sz
img, coords = pad_and_or_crop(img, self.input_sz, mode="random")
label, _ = pad_and_or_crop(label, self.input_sz, mode="fixed",
coords=coords)
_, mask_img1 = self._filter_label(label)
# uint8 tensor as masks should be binary, also for consistency with
# prepare_train, but converted to float32 in main loop because is used
# multiplicatively in loss
mask_img1 = torch.from_numpy(mask_img1.astype(np.uint8)).cuda()
# make img2 different from img1 (img)
# tf_mat can be:
# *A, from img2 to img1 (will be applied to img2's heatmap)-> img1 space
# input img1 tf: *tf.functional or pil.image
# input mask tf: *none
# output heatmap: *tf.functional (parallel), inverse of what is used
# for inputs, create inverse of this tf in [-1, 1] format
# B, from img1 to img2 (will be applied to img1's heatmap)-> img2 space
# input img1 tf: pil.image
# input mask tf: pil.image (discrete)
# output heatmap: tf.functional, create copy of this tf in [-1,1] format
# tf.function tf_mat: translation is opposite to what we'd expect (+ve 1
# is shift half towards left)
# but rotation is correct (-sin in top right = counter clockwise)
# flip is [[-1, 0, 0], [0, 1, 0], [0, 0, 1]]
# img2 = flip(affine1_to_2(img1))
# => img1_space = affine1_to_2^-1(flip^-1(img2_space))
# = affine2_to_1(flip^-1(img2_space))
# so tf_mat_img2_to_1 = affine2_to_1 * flip^-1 (order matters as not diag)
# flip^-1 = flip
# no need to tf label, as we're doing option A, mask needed in img1 space
# converting to PIL does not change underlying np datatype it seems
img1 = Image.fromarray(img.astype(np.uint8))
# (img2) do jitter, no tf_mat change
img2 = self.jitter_tf(img1) # not in place, new memory
img1 = np.array(img1)
img2 = np.array(img2)
# channels still last
if not self.no_sobel:
img1 = custom_greyscale_numpy(img1, include_rgb=self.include_rgb)
img2 = custom_greyscale_numpy(img2, include_rgb=self.include_rgb)
img1 = img1.astype(np.float32) / 255.
img2 = img2.astype(np.float32) / 255.
# convert both to channel-first tensor format
# make them all cuda tensors now, except label, for optimality
img1 = torch.from_numpy(img1).permute(2, 0, 1).cuda()
img2 = torch.from_numpy(img2).permute(2, 0, 1).cuda()
# mask if required
if self.mask_input:
masked = 1 - mask_img1
img1[:, masked] = 0
img2[:, masked] = 0
# (img2) do affine if nec, tf_mat changes
if self.use_random_affine:
affine_kwargs = {"min_rot": self.aff_min_rot, "max_rot": self.aff_max_rot,
"min_shear": self.aff_min_shear,
"max_shear": self.aff_max_shear,
"min_scale": self.aff_min_scale,
"max_scale": self.aff_max_scale}
img2, affine1_to_2, affine2_to_1 = random_affine(img2,
**affine_kwargs) #
# tensors
else:
affine2_to_1 = torch.zeros([2, 3]).to(torch.float32).cuda() # identity
affine2_to_1[0, 0] = 1
affine2_to_1[1, 1] = 1
# (img2) do random flip, tf_mat changes
if np.random.rand() > self.flip_p:
img2 = torch.flip(img2, dims=[2]) # horizontal, along width
# applied affine, then flip, new = flip * affine * coord
# (flip * affine)^-1 is just flip^-1 * affine^-1.
# No order swap, unlike functions...
# hence top row is negated
affine2_to_1[0, :] *= -1.
if RENDER_DATA:
render(img1, mode="image", name=("train_data_img1_%d" % index))
render(img2, mode="image", name=("train_data_img2_%d" % index))
render(affine2_to_1, mode="matrix",
name=("train_data_affine2to1_%d" % index))
render(mask_img1, mode="mask", name=("train_data_mask_%d" % index))
return img1, img2, affine2_to_1, mask_img1
def _prepare_train_single(self, index, img, label):
# Returns one pair only, i.e. without transformed second image.
# Used for standard CNN training (baselines).
# This returns gpu tensors.
# label is passed in canonical [0 ... 181] indexing
assert (img.shape[:2] == label.shape)
img = img.astype(np.float32)
label = label.astype(np.int32)
# shrink original images, for memory purposes, otherwise no point
if self.pre_scale_all:
assert (self.pre_scale_factor < 1.)
img = cv2.resize(img, dsize=None, fx=self.pre_scale_factor,
fy=self.pre_scale_factor,
interpolation=cv2.INTER_LINEAR)
label = cv2.resize(label, dsize=None, fx=self.pre_scale_factor,
fy=self.pre_scale_factor,
interpolation=cv2.INTER_NEAREST)
if self.use_random_scale:
# bilinear interp requires float img
scale_factor = (np.random.rand() * (self.scale_max - self.scale_min)) + \
self.scale_min
img = cv2.resize(img, dsize=None, fx=scale_factor, fy=scale_factor,
interpolation=cv2.INTER_LINEAR)
label = cv2.resize(label, dsize=None, fx=scale_factor, fy=scale_factor,
interpolation=cv2.INTER_NEAREST)
# random crop to input sz
img, coords = pad_and_or_crop(img, self.input_sz, mode="random")
label, _ = pad_and_or_crop(label, self.input_sz, mode="fixed",
coords=coords)
_, mask_img1 = self._filter_label(label)
# uint8 tensor as masks should be binary, also for consistency with
# prepare_train, but converted to float32 in main loop because is used
# multiplicatively in loss
mask_img1 = torch.from_numpy(mask_img1.astype(np.uint8)).cuda()
# converting to PIL does not change underlying np datatype it seems
img1 = Image.fromarray(img.astype(np.uint8))
img1 = self.jitter_tf(img1) # not in place, new memory
img1 = np.array(img1)
# channels still last
if not self.no_sobel:
img1 = custom_greyscale_numpy(img1, include_rgb=self.include_rgb)
img1 = img1.astype(np.float32) / 255.
# convert both to channel-first tensor format
# make them all cuda tensors now, except label, for optimality
img1 = torch.from_numpy(img1).permute(2, 0, 1).cuda()
# mask if required
if self.mask_input:
masked = 1 - mask_img1
img1[:, masked] = 0
if self.use_random_affine:
affine_kwargs = {"min_rot": self.aff_min_rot, "max_rot": self.aff_max_rot,
"min_shear": self.aff_min_shear,
"max_shear": self.aff_max_shear,
"min_scale": self.aff_min_scale,
"max_scale": self.aff_max_scale}
img1, _, _ = random_affine(img1, **affine_kwargs) # tensors
if np.random.rand() > self.flip_p:
img1 = torch.flip(img1, dims=[2]) # horizontal, along width
if RENDER_DATA:
render(img1, mode="image", name=("train_data_img1_%d" % index))
render(mask_img1, mode="mask", name=("train_data_mask_%d" % index))
return img1, mask_img1
def _prepare_test(self, index, img, label):
# This returns cpu tensors.
# Image: 3D with channels last, float32, in range [0, 1] (normally done
# by ToTensor).
# Label map: 2D, flat int64, [0 ... sef.gt_k - 1]
# label is passed in canonical [0 ... 181] indexing
assert (img.shape[:2] == label.shape)
img = img.astype(np.float32)
label = label.astype(np.int32)
# shrink original images, for memory purposes, otherwise no point
if self.pre_scale_all:
assert (self.pre_scale_factor < 1.)
img = cv2.resize(img, dsize=None, fx=self.pre_scale_factor,
fy=self.pre_scale_factor,
interpolation=cv2.INTER_LINEAR)
label = cv2.resize(label, dsize=None, fx=self.pre_scale_factor,
fy=self.pre_scale_factor,
interpolation=cv2.INTER_NEAREST)
# center crop to input sz
img, _ = pad_and_or_crop(img, self.input_sz, mode="centre")
label, _ = pad_and_or_crop(label, self.input_sz, mode="centre")
# finish
if not self.no_sobel:
img = custom_greyscale_numpy(img, include_rgb=self.include_rgb)
img = img.astype(np.float32) / 255.
img = torch.from_numpy(img).permute(2, 0, 1)
if RENDER_DATA:
render(label, mode="label", name=("test_data_label_pre_%d" % index))
# convert to coarse if required, reindex to [0, gt_k -1], and get mask
label, mask = self._filter_label(label)
# mask if required
if self.mask_input:
masked = 1 - mask
img[:, masked] = 0
if RENDER_DATA:
render(img, mode="image", name=("test_data_img_%d" % index))
render(label, mode="label", name=("test_data_label_post_%d" % index))
render(mask, mode="mask", name=("test_data_mask_%d" % index))
# dataloader must return tensors (conversion forced in their code anyway)
return img, torch.from_numpy(label), torch.from_numpy(mask.astype(np.uint8))
def __getitem__(self, index):
image_id = self.files[index]
image, label = self._load_data(image_id)
if self.purpose == "train":
if not self.single_mode:
return self._prepare_train(index, image, label)
else:
return self._prepare_train_single(index, image, label)
else:
assert (self.purpose == "test")
return self._prepare_test(index, image, label)
def __len__(self):
return len(self.files)
def _check_gt_k(self):
raise NotImplementedError()
def _filter_label(self):
raise NotImplementedError()
def _set_files(self):
raise NotImplementedError()
def _load_data(self, image_id):
raise NotImplementedError()
# ------------------------------------------------------------------------------
# Handles which images are eligible
class _Coco10k(_Coco):
"""Base class
This contains fields and methods common to all COCO 10k datasets:
(COCO-fine) (182)
COCO-coarse (27)
COCO-few (6)
(COCOStuff-fine) (91)
COCOStuff-coarse (15)
COCOStuff-few (3)
"""
def __init__(self, **kwargs):
super(_Coco10k, self).__init__(**kwargs)
self._set_files()
def _set_files(self):
if self.split in ["train", "test", "all"]:
# deterministic order - important - so >1 dataloader actually meaningful
file_list = osp.join(self.root, "imageLists", self.split + ".txt")
file_list = tuple(open(file_list, "r"))
file_list = [id_.rstrip() for id_ in file_list]
self.files = file_list
else:
raise ValueError("Invalid split name: {}".format(self.split))
def _load_data(self, image_id):
image_path = osp.join(self.root, "images", image_id + ".jpg")
label_path = osp.join(self.root, "annotations", image_id + ".mat")
image = cv2.imread(image_path, cv2.IMREAD_COLOR).astype(np.uint8)
label = sio.loadmat(label_path)["S"].astype(np.int32) # [0, 182]
label -= 1 # unlabeled (0 -> -1)
# label should now be [-1, 0 ... 181], 91 each
return image, label
class _Coco164k(_Coco):
"""Base class
This contains fields and methods common to all COCO 164k datasets
This is too huge to train in reasonable time
"""
def __init__(self, **kwargs):
super(_Coco164k, self).__init__(**kwargs)
self._set_files()
def _set_files(self):
# Create data list by parsing the "images" folder
if self.split in ["train2017", "val2017"]:
file_list = sorted(
glob(osp.join(self.root, "images", self.split, "*.jpg")))
file_list = [f.split("/")[-1].replace(".jpg", "") for f in file_list]
self.files = file_list
else:
raise ValueError("Invalid split name: {}".format(self.split))
def _load_data(self, image_id):
# Set paths
image_path = osp.join(self.root, "images", self.split, image_id + ".jpg")
label_path = osp.join(self.root, "annotations", self.split,
image_id + ".png")
# Load an image
image = cv2.imread(image_path, cv2.IMREAD_COLOR).astype(np.uint8)
label = cv2.imread(label_path, cv2.IMREAD_GRAYSCALE).astype(np.int32)
label[label == 255] = -1 # to be consistent with 10k
return image, label
class _Coco164kCuratedFew(_Coco):
"""Base class
This contains fields and methods common to all COCO 164k curated few datasets:
(curated) Coco164kFew_Stuff
(curated) Coco164kFew_Stuff_People
(curated) Coco164kFew_Stuff_Animals
(curated) Coco164kFew_Stuff_People_Animals
"""
def __init__(self, **kwargs):
super(_Coco164kCuratedFew, self).__init__(**kwargs)
# work out name
config = kwargs["config"]
assert (config.use_coarse_labels) # we only deal with coarse labels
self.include_things_labels = config.include_things_labels # people
self.incl_animal_things = config.incl_animal_things # animals
version = config.coco_164k_curated_version
name = "Coco164kFew_Stuff"
if self.include_things_labels and self.incl_animal_things:
name += "_People_Animals"
elif self.include_things_labels:
name += "_People"
elif self.incl_animal_things:
name += "_Animals"
self.name = (name + "_%d" % version)
print("Specific type of _Coco164kCuratedFew dataset: %s" % self.name)
self._set_files()
def _set_files(self):
# Create data list by parsing the "images" folder
if self.split in ["train2017", "val2017"]:
file_list = osp.join(self.root, "curated", self.split, self.name + ".txt")
file_list = tuple(open(file_list, "r"))
file_list = [id_.rstrip() for id_ in file_list]
self.files = file_list
else:
raise ValueError("Invalid split name: {}".format(self.split))
def _load_data(self, image_id):
# same as _Coco164k
# Set paths
image_path = osp.join(self.root, "images", self.split, image_id + ".jpg")
label_path = osp.join(self.root, "annotations", self.split,
image_id + ".png")
# Load an image
image = cv2.imread(image_path, cv2.IMREAD_COLOR).astype(np.uint8)
label = cv2.imread(label_path, cv2.IMREAD_GRAYSCALE).astype(np.int32)
label[label == 255] = -1 # to be consistent with 10k
return image, label
class _Coco164kCuratedFull(_Coco):
"""Base class
This contains fields and methods common to all COCO 164k curated full
datasets:
(curated) Coco164kFull_Stuff_Coarse
"""
def __init__(self, **kwargs):
super(_Coco164kCuratedFull, self).__init__(**kwargs)
# work out name
config = kwargs["config"]
assert (config.use_coarse_labels) # we only deal with coarse labels
assert (not config.include_things_labels)
assert (not config.incl_animal_things)
version = config.coco_164k_curated_version
self.name = "Coco164kFull_Stuff_Coarse_%d" % version
print("Specific type of _Coco164kCuratedFull dataset: %s" % self.name)
self._set_files()
def _set_files(self):
# Create data list by parsing the "images" folder
if self.split in ["train2017", "val2017"]:
file_list = osp.join(self.root, "curated", self.split,
self.name + ".txt")
file_list = tuple(open(file_list, "r"))
file_list = [id_.rstrip() for id_ in file_list]
self.files = file_list
else:
raise ValueError("Invalid split name: {}".format(self.split))
def _load_data(self, image_id):
# same as _Coco164k
# Set paths
image_path = osp.join(self.root, "images", self.split, image_id + ".jpg")
label_path = osp.join(self.root, "annotations", self.split,
image_id + ".png")
# Load an image
image = cv2.imread(image_path, cv2.IMREAD_COLOR).astype(np.uint8)
label = cv2.imread(label_path, cv2.IMREAD_GRAYSCALE).astype(np.int32)
label[label == 255] = -1 # to be consistent with 10k
return image, label
# ------------------------------------------------------------------------------
# Handles Full vs Few
class _CocoFull(_Coco):
"""
This contains methods for the following datasets
(Full = original labels, coarse or fine)
(COCO-fine) (182)
COCO-coarse (27)
(COCOStuff-fine) (91)
COCOStuff-coarse (15)
"""
def __init__(self, **kwargs):
super(_CocoFull, self).__init__(**kwargs)
config = kwargs["config"]
# if coarse, index corresponds to order in cocostuff_fine_to_coarse.py
self.use_coarse_labels = config.use_coarse_labels
self.include_things_labels = config.include_things_labels
self._check_gt_k()
def _fine_to_coarse(self, label_map):
# label_map is in fine indexing
# can't be in place!
new_label_map = np.zeros(label_map.shape, dtype=label_map.dtype)
# -1 stays -1
for c in xrange(182):
new_label_map[label_map == c] = self._fine_to_coarse_dict[c]
return new_label_map
def _check_gt_k(self):
if self.use_coarse_labels:
if self.include_things_labels:
assert (self.gt_k == 27)
else:
assert (self.gt_k == 15)
else:
if self.include_things_labels:
assert (self.gt_k == 182)
else:
assert (self.gt_k == 91)
def _filter_label(self, label):
# expects np array in fine labels ([0, 181]) and returns np arrays
# convert to coarse if required, and reindex to [0, gt_k -1], and get mask
# do we care about what is in masked portion of label map - no
# in eval, mask used to select, others ignored
# things: 91 classes (0-90), 12 superclasses (0-11)
# stuff: 91 classes (91-181), 15 superclasses (12-26)
if self.use_coarse_labels:
label = self._fine_to_coarse(label)
if self.include_things_labels:
first_allowed_index = 0
else:
first_allowed_index = 12 # first coarse stuff index
else:
if self.include_things_labels:
first_allowed_index = 0
else:
first_allowed_index = 91 # first fine stuff index
# always excludes unlabelled (<= -1)
mask = (label >= first_allowed_index)
assert (mask.dtype == np.bool)
# put in [0, gt_k], gt_k can be 27, 15, 182, 91
label -= first_allowed_index
return label, mask
class _CocoFew(_Coco):
"""
This contains methods for the following datasets
COCO-few (6)
COCOStuff-few (3)
"""
def __init__(self, **kwargs):
super(_CocoFew, self).__init__(**kwargs)
config = kwargs["config"]
assert (config.use_coarse_labels) # we only deal with coarse labels
self.include_things_labels = config.include_things_labels
self.incl_animal_things = config.incl_animal_things
self._check_gt_k()
# indexes correspond to order in these lists
self.label_names = [
"sky-stuff",
"plant-stuff",
"ground-stuff",
]
# CHANGED. Can have animals and/or people.
if self.include_things_labels:
self.label_names += ["person-things"]
if self.incl_animal_things:
self.label_names += ["animal-things"]
assert (len(self.label_names) == self.gt_k)
# make dict that maps fine labels to our labels
self._fine_to_few_dict = self._make_fine_to_few_dict()
def _make_fine_to_few_dict(self):
# only make indices
self.label_orig_coarse_inds = []
for label_name in self.label_names:
orig_coarse_ind = cocostuff_fine_to_coarse._sorted_coarse_names.index(
label_name)
self.label_orig_coarse_inds.append(orig_coarse_ind)
print("label_orig_coarse_inds for this dataset: ")
print(self.label_orig_coarse_inds)
# excludes -1 (fine - see usage in filter label - as with Coco10kFull)
_fine_to_few_dict = {}
for c in xrange(182):
orig_coarse_ind = self._fine_to_coarse_dict[c]
if orig_coarse_ind in self.label_orig_coarse_inds:
new_few_ind = self.label_orig_coarse_inds.index(orig_coarse_ind)
# print("assigning fine %d coarse %d to new ind %d" % (c,
# orig_coarse_ind,
# new_few_ind))
else:
new_few_ind = -1
_fine_to_few_dict[c] = new_few_ind
# print("fine to few dict:")
# print(_fine_to_few_dict)
return _fine_to_few_dict
def _check_gt_k(self):
# Can have animals and/or people.
expected_gt_k = 3
if self.include_things_labels:
expected_gt_k += 1
if self.incl_animal_things:
expected_gt_k += 1
assert (self.gt_k == expected_gt_k)
def _filter_label(self, label):
# expects np array in fine labels ([-1, 181]) and returns np arrays
# use coarse labels, reindex to [-1, gt_k -1], and get mask
# do we care about what is in masked portion of label map - no
# in eval, mask used to select, others ignored
# min = label.min()
# max = label.max()
# if min < -1:
# print("smaller than expected %d" % min)
# assert(False)
# if max >= 182:
# print("bigger than expected %d" % max)
# assert(False)
# can't be in place!
# -1 stays -1
new_label_map = np.zeros(label.shape, dtype=label.dtype)
for c in xrange(182):
new_label_map[label == c] = self._fine_to_few_dict[c]
mask = (new_label_map >= 0)
assert (mask.dtype == np.bool)
return new_label_map, mask
# ------------------------------------------------------------------------------
# All 4 combinations of 10k-164k, Full-Few (Full includes coarse or fine)
class Coco10kFull(_Coco10k, _CocoFull):
def __init__(self, **kwargs):
super(Coco10kFull, self).__init__(**kwargs)
class Coco10kFew(_Coco10k, _CocoFew):
def __init__(self, **kwargs):
super(Coco10kFew, self).__init__(**kwargs)
class Coco164kFull(_Coco164k, _CocoFull):
def __init__(self, **kwargs):
super(Coco164kFull, self).__init__(**kwargs)
class Coco164kFew(_Coco164k, _CocoFew):
def __init__(self, **kwargs):
super(Coco164kFew, self).__init__(**kwargs)
# Only 2 top level class options for curated datasets
class Coco164kCuratedFew(_Coco164kCuratedFew, _CocoFew):
def __init__(self, **kwargs):
super(Coco164kCuratedFew, self).__init__(**kwargs)
class Coco164kCuratedFull(_Coco164kCuratedFull, _CocoFull):
def __init__(self, **kwargs):
super(Coco164kCuratedFull, self).__init__(**kwargs)