-
Notifications
You must be signed in to change notification settings - Fork 5
/
pre_process_dataset.py
1191 lines (979 loc) · 51 KB
/
pre_process_dataset.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
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
import os
import numpy as np
import shutil
#from os import path as osp
import cv2
import lmdb
import sys
from multiprocessing import Pool
from os import path as osp
from tqdm import tqdm
from utils.ws_ssim import ws_ssim
from PIL import Image
########################
##### Dataset path #####
########################
org_dir = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_lc_LR240' # lc 原始数据库ERP格式降低分辨率后的YUV 480 240, 然后ERP 转换成三种不同格式的 180个
imp_dir = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_lc_imp_ref240' # 各种格式的ref视频HEVC压缩后的视频 540个 还没有经过重映射回ERP格式
#imp_erp_dir = r'/media/yl/yl_8t/data_vqa_yl720'
imp_erp_dir = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_yl240' # 经过两种失真后的540个ODV
com_cmp_dir = r'/media/yl/yl_8t/com_cmp'
dataset_utils = r'/home/yl/bvqa360/data/' # 存放各种文件信息 dmos name等txt
orig_data = r'/media/liutie/pami_yl/DatabaseVideoYUV' # LC 大硬盘里的最原始的数据库 yuv
lr_data = r'/media/liutie/pami_yl/data_vqa_lc_LR' # 处理第一步
host_dir = r'/media/yl/yl_8t/'
log_dir = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_lc_logbin240/log/' # H.265压缩编码后信息储存的地方
bin_dir = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_lc_logbin240/bin/'
#########################################################################################################################################################
##### Create the lc ODV-VQA dataset (format: YUV; number: 540; resolution: (480, 240); Projection: ERP, RCMP, TSP; Compression levels: QP=27,37,42) #####
#########################################################################################################################################################
# 1. 原始ODV降分辨率： 将60个原始ODV (yuv格式，orig_data/Reference目录下：10个Group, 每个Group里面有6个原始无失真参考ODV) 用ffmpeg降低分辨率到480×240，
# 放入到 lr_data 目录下 （另一个8T lc 硬盘 中），用SCP命令将所有将分辨率后的ODV传送到我的硬盘 （org_dir/Reference 目录下）。
def org_lr():
for group in os.listdir(orig_data):
if not os.path.exists(os.path.join(lr_data, group)):
os.mkdir(os.path.join(lr_data, group))
for obj in os.listdir(os.path.join(orig_data, group)):
#print(obj)
if os.path.isdir(os.path.join(orig_data, group, obj)): # reference directory
print("it's a directory {}".format(obj))
if not os.path.exists(os.path.join(lr_data, group, obj)):
os.mkdir(os.path.join(lr_data, group, obj))
for ref in os.listdir(os.path.join(orig_data, group, obj)):
res = ref.split('_')[1]
res_lr = '480x240'
org_ref_video = os.path.join(orig_data, group, obj, ref)
lr_ref_video = os.path.join(lr_data, group, obj, ref)
cmd1 = r'ffmpeg -f rawvideo -s {} -pix_fmt yuv420p -i {} -s {} -c:v rawvideo {}'.format(res, org_ref_video, res_lr, lr_ref_video)
os.system(cmd1)
else:
print('it is a special file(socket,FIFO,device file)')
# 2. 加入映射失真： 将所有60个原始参考ODV 使用360tools转化为CMP TSP ERP三种格式，共3*60=180个，每组18个，还放在org_dir目录下；
# 执行程序：将.py置于/home/yl/360tools/bin 下进行映射转换。
def convert_360():
for group in os.listdir(org_dir): # Group1...
if not os.path.exists(os.path.join(imp_dir, group)):
os.mkdir(os.path.join(imp_dir, group))
# if not os.path.exists(os.path.join(org_dir, group, 'CMP')):
# os.mkdir(os.path.join(org_dir, group, 'CMP'))
ref_dir = os.path.join(org_dir, group, 'Reference')
for ref_video in os.listdir(ref_dir):
ref_video_sp = ref_video.split('_')
erp_name = '_'.join([ref_video_sp[0], 'ERP']+ref_video_sp[1:])
tsp_name = '_'.join([ref_video_sp[0], 'TSP']+ref_video_sp[1:])
cmp_name = '_'.join([ref_video_sp[0], 'RCMP']+ref_video_sp[1:])
#rcmp_name = '_'.join([ref_video_sp[0], 'RCMP']+ref_video_sp[1:])
ref_video_pth = os.path.join(ref_dir, ref_video)
erp_video_pth = os.path.join(os.path.join(org_dir, group), erp_name)
tsp_video_pth = os.path.join(os.path.join(org_dir, group), tsp_name)
cmp_video_pth = os.path.join(os.path.join(org_dir, group), cmp_name)
#rcmp_video_pth = os.path.join(os.path.join(org_dir, group), rcmp_name)
cmd1 = './360tools_conv -i {} -w 480 -h 240 -x 1 -o {} -l 480 -m 240 -y 1 -f 7'.format(ref_video_pth, tsp_video_pth)
cmd2 = './360tools_conv -i {} -w 480 -h 240 -x 1 -o {} -l 480 -m 360 -y 1 -f 3'.format(ref_video_pth, cmp_video_pth)
#cmd3 = './360tools_conv -i {} -w 1280 -h 960 -x 1 -o {} -l 1280 -m 960 -y 1 -f 13'.format(cmp_video_pth, rcmp_video_pth)
shutil.copyfile(ref_video_pth, erp_video_pth)
os.system(cmd1)
os.system(cmd2)
#os.system(cmd3)
# 3. 加入压缩失真：将180个已经加入映射失真的ODV每一个都加入QP=27, 37, 42的压缩失真， 存放在 imp_dir 目录下。 10个Group,每个Group 54个ODV。
# 新的 480×240 ODV, 180个 ERP TSP RCMP失真的所有yuv编成.sh
# 多线程编码：目录在 /media/yl/yl_8t/Encoder_exdata_VQA_180Seqs_RA_Schedule
# 首先删掉Encoder_#1-16 所有目录，改写run.info文件，使他成为0 1 0 （视频全部编码好就成为540 1 0）
# 然后把rename后的.sh文件重命名为run.sh （本程序：vqa_lc_rename240.txt-> run.sh）
# 最后命令行跑 python run_multi_threads.py 开始多线程编码
def comp_vqa_lc_HM_rename_240():
# 写之前，先检验文件是否存在，存在就删掉
if os.path.exists(os.path.join(host_dir, 'vqa_lc_rename240.txt')):
os.remove(os.path.join(host_dir, 'vqa_lc_rename240.txt'))
num_frames = np.loadtxt(os.path.join(dataset_utils, 'vqa_frameNums.txt'))
num_frames = num_frames.reshape((10, 18))
f = open(os.path.join(host_dir, 'vqa_lc_rename240.txt'), 'w')
g = -1
for gg in range(10):
group = 'Group' + str(gg + 1)
if not os.path.exists(os.path.join(log_dir, group)):
os.mkdir(os.path.join(log_dir, group))
if not os.path.exists(os.path.join(bin_dir, group)):
os.mkdir(os.path.join(bin_dir, group))
g += 1
num_group_frame = num_frames[g]
index = -1
for obj in os.listdir(os.path.join(org_dir, group)):
if os.path.isfile(os.path.join(org_dir, group, obj)):
if obj[0] == 'G':
index += 1
obj_splt = obj.split('_')
obj_proj = obj_splt[1]
if obj_proj == 'ERP':
W, H = 480, 240
elif obj_proj == 'RCMP':
W, H = 480, 360
elif obj_proj == 'TSP':
W, H = 480, 240
nf = num_group_frame[index]
org_file = os.path.join(org_dir, group, obj)
obj_name_splt = obj.split('.')[0:-1]
obj_name_splt = '.'.join(obj_name_splt)
print(obj_name_splt)
for qp_num in [27, 37, 42]:
print(obj_name_splt)
qp_num_str = 'qp' + str(qp_num)
obj_yuv = obj_name_splt + '_' + qp_num_str + '.yuv'
obj_log = obj_name_splt + '_' + qp_num_str + '.txt'
obj_bin = obj_name_splt + '_' + qp_num_str + '.bin'
# print(obj_yuv)
com_file = os.path.join(imp_dir, group, obj_yuv)
log_file = os.path.join(log_dir, group, obj_log)
bin_file = os.path.join(bin_dir, group, obj_bin)
vqa_coding_cmd = './TAppEncoderStatic -c encoder_randomaccess_main.cfg -c encoder_others.cfg -i {} -wdt {} -hgt {} -fr 30 -f {} -q {} -b {} -o {} >{}\n'.format(
org_file, W, H, int(nf), qp_num, bin_file, com_file, log_file)
f.writelines(vqa_coding_cmd)
else:
continue
f.close()
# 4. 重映射： 将压缩后的540个ODV重新映射回ERP格式，将所有540个ODV分成10组放入到 imp_erp_dir 目录中。
# 运行程序：将.py置于/home/yl/360tools/bin 下进行映射转换。
def com_yuv_convert_yl240():
f = open(os.path.join(dataset_utils, 'train.txt'),"r")
lines = f.readlines() #读取全部内容 ，并以列表方式返回
namelist = []
for num_ref in range(0, 540):
line = lines[num_ref]
line_list = line.split() # string to list
name = line_list[3]
filename = os.path.splitext(name.split(os.sep)[-1])[0] # 去路径，去后缀，G1AbandonedKingdom_ERP_7680x3840_fps30_qp27_45406k
namelist += [filename]
print(namelist)
for vid in range(0, 540):
vid_name = namelist[vid]
vid_name_spl = vid_name.split('_')
group_index = vid_name_spl[0][1]
group_name = 'Group' + group_index
if vid>=(540-54):
group_name = 'Group' + str(10)
if not os.path.exists(os.path.join(imp_erp_dir, group_name)):
os.mkdir(os.path.join(imp_erp_dir, group_name))
vid_pth = os.path.join(imp_dir, group_name)
if vid_name_spl[1] == 'ERP':
vid_org_name = ('_').join(vid_name_spl[:-1])
vid_orh_pth = os.path.join(vid_pth, vid_org_name+'.yuv')
vid_out_pth = os.path.join(imp_erp_dir, group_name, vid_name+'.yuv')
shutil.copyfile(vid_orh_pth, vid_out_pth)
print('Successfully !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!')
elif vid_name_spl[1] == 'TSP':
vid_org_name = ('_').join(vid_name_spl[:-2])
vid_orh_pth = os.path.join(vid_pth, vid_org_name+'.yuv')
vid_out_pth = os.path.join(imp_erp_dir, group_name, vid_name+'.yuv')
cmd1 = './360tools_conv -i {} -w 480 -h 240 -x 1 -o {} -l 480 -m 240 -y 1 -f 8'.format(vid_orh_pth, vid_out_pth)
os.system(cmd1)
print('Successfully !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!')
elif vid_name_spl[1] == 'RCMP':
vid_org_name = ('_').join(vid_name_spl[:-2])
vid_orh_pth = os.path.join(vid_pth, vid_org_name+'.yuv')
vid_out_pth = os.path.join(imp_erp_dir, group_name, vid_name+'.yuv')
cmd2 = './360tools_conv -i {} -w 480 -h 360 -x 1 -o {} -l 480 -m 240 -y 1 -f 4'.format(vid_orh_pth, vid_out_pth)
os.system(cmd2)
print('Successfully !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!')
else:
continue
########## Apendix of 360tools ##############
# ./360tools_conv
# -i argument should be set
# < Usage >
# -c, --config [STRING] (optional)
# : config file name
# -i, --input [STRING]
# : input file name
# -o, --output [STRING]
# : output file name
# -w, --width [INTEGER]
# : width of input image
# -h, --height [INTEGER]
# : height of input image
# -f, --convfmt [INTEGER]
# : converting format
# 0: ERP to CPP
# 1: ERP to ISP
# 2: ISP to ERP
# 3: ERP to CMP
# 4: CMP to ERP
# 5: ERP to OHP
# 6: OHP to ERP
# 7: ERP to TSP
# 8: TSP to ERP
# 9: ERP to SSP
# 10: SSP to ERP
# 11: ISP to RISP
# 12: RISP to ISP
# 13: CMP to RCMP
# 14: RCMP to CMP
# 15: OHP to ROHP
# 16: ROHP to OHP
# 21: ERP to RISP
# 22: RISP to ERP
# 25: ERP to COHP
# 26: COHP to ERP
# 31: CPP to ERP
# 32: CPP to ISP
# 33: CPP to CMP
# 34: CPP to OHP
# 35: CPP to TSP
# 36: CPP to SSP
#
# -l, --out_width [INTEGER]
# : width of output image. if not set, this is same value with width of input image
# -m, --out_height [INTEGER]
# : height of output image. if not set, this is same value with height of input image
# -n, --frmnum [INTEGER] (optional)
# : number of frames to be converted
# -a, --align [INTEGER] (optional)
# : vertical align
# -u, --no_pad [FLAG] (optional)
# : turn off padding
# -x, --cs_in [INTEGER] (optional)
# : Input Color Space
# 1: YUV420 8-bit
# 2: YUV420 10-bit
# -y, --cs_out [INTEGER] (optional)
# : Output Color Space
# 1: YUV420 8-bit
# 2: YUV420 10-bit
# -p, --pitch [INTEGER] (optional)
# : TSP pitch angle [0,180] default is 90
# -t, --yaw [INTEGER] (optional)
# : TSP yaw angle [0,360] default is 0
#########################################################################################################################################################
##### Prepare the train (LMDB) and test (PNG) set for BVQA360 #####
#########################################################################################################################################################
####################################
##### Dataset path for BVQA360 #####
####################################
dataset_240= r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_yl240' # 10组视频 每组54个失真的 共540个
png240_dir = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_yl240_png' # 所有540个目录 每个目录有n帧png图像 每个目录帧数不一样
odv_train240_png = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_yl240_png_train' # train 训练集 432个目录的png
odv_test_png240 = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_yl240_png_test' # test 测试集 108个目录的png
new_data_png240 = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_all_test_png' # 所有测试的视频所有帧全部放在这一个目录下---》这个是用于test时候的输入图像
odv_ref_png240 = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_yl240_png_ref' # 十个Group, 每个Group中有6个reference ODV目录，每个目录下有n帧 PNG图像
odvLmdb_train_pth2 = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_yl240_lmdb_train/odv240-vqa_dataset_train.lmdb'
# 这4个文件都是按照train 432 test 108 共540个 顺序排列的 存放视频信息 包括： 名字 dmos 帧数 w h num_frame
video_trates_info = r'/home/yl/bvqa360/data/train_test_240ODV_all_info.txt' # 0 G1AbandonedKingdom_ERP_7680x3840_fps30_qp27_45406k 0.414890202945909 300 240 480
video_tratest_NSTSS_info = r'/home/yl/video_file_names_odv240.txt' # G1AbandonedKingdom_ERP_7680x3840_fps30_qp27_45406k 480 240 300
video_tratest_dmos = r'/home/yl/bvqa360/data/train_test_240ODV_dmos.txt' # 0.414890202945909, 0.490389955570886, 0.566788675185082, ..., 0.620770373788080
video_tratest_name = r'/home/yl/bvqa360/data/train_test_240ODV_name.txt' # G1AbandonedKingdom_ERP_7680x3840_fps30_qp27_45406k, G1AbandonedKingdom_ERP_7680x3840_fps30_qp37_9283k,...,
# ID是reference 序号 0~59
odv_train_id = [0, 1 ,2, 5 ,6 ,7, 8 ,9, 10, 11, 13, 14, 15, 16, 18, 20, 22, 24, 25, 26, 27, 29, 31 ,32, 33, 34, 35, 36, 37, 38, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 59]
odv_test_id = [3, 4 ,12, 17, 19 ,21 ,23 ,28, 30, 39, 40, 58]
def scandir(dir_path, suffix=None, recursive=False, full_path=False):
"""Scan a directory to find the interested files.
suffix: 搜索文件的后缀 recursive：递归搜索， full_path: 搜索包括目录
# 在dir_pth目录下 可以递归或者不递归的生成一个文件迭代器，包含所有文件名，不包含目录。
Args:
dir_path (str): Path of the directory.
suffix (str | tuple(str), optional): File suffix that we are
interested in. Default: None.
recursive (bool, optional): If set to True, recursively scan the
directory. Default: False. 递归查找文件或者目录
full_path (bool, optional): If set to True, include the dir_path.
Default: False.
Returns:
A generator for all the interested files with relative pathes.
"""
if (suffix is not None) and not isinstance(suffix, (str, tuple)):
raise TypeError('"suffix" must be a string or tuple of strings')
root = dir_path
def _scandir(dir_path, suffix, recursive):
for entry in os.scandir(dir_path): # generator 可迭代生成器
#print(entry) # <DirEntry 'meta_info_REDS4_test_GT.txt'>
#print(entry.name) # meta_info_REDS4_test_GT.txt
if not entry.name.startswith('.') and entry.is_file():
if full_path:
return_path = entry.path
else:
#print(entry.path) # /home/yl/bvqa360/basicsr/data/transforms.py
return_path = osp.relpath(entry.path, root) # os.path.relpath()Python中的方法用于从当前工作目录或给定目录获取到给定路径的相对文件路径。
#print(return_path) # transforms.py
if suffix is None:
yield return_path
elif return_path.endswith(suffix):
yield return_path
else:
if recursive:
yield from _scandir(
entry.path, suffix=suffix, recursive=recursive)
else:
continue
return _scandir(dir_path, suffix=suffix, recursive=recursive)
# 1. 将 dataset_240 .yuv 视频通过ffmpeg转换为 .png 图像；png240_dir 包含540个目录，每个目录中有每一帧的png图像。
def yuv2png_240():
for group in os.listdir(dataset_240):
group_pth = os.path.join(dataset_240, group)
for vid in os.listdir(group_pth):
name, ext = os.path.splitext(vid)
vid_pth = os.path.join(dataset_240, group, vid)
if not os.path.exists(os.path.join(png240_dir, name)):
os.mkdir(os.path.join(png240_dir, name))
out_dir = os.path.join(png240_dir, name, '%3d.png')
cmd = 'ffmpeg -f rawvideo -pixel_format yuv420p -s 480x240 -i {} {}'.format(vid_pth, out_dir)
os.system(cmd)
def yuv2png_ref_240():
for group in os.listdir(org_dir):
if not os.path.exists(os.path.join(odv_ref_png240, group)):
os.mkdir(os.path.join(odv_ref_png240, group))
for obj in os.listdir(os.path.join(org_dir, group)):
#print(obj)
if os.path.isdir(os.path.join(org_dir, group, obj)): # reference directory
print("it's a directory {}".format(obj))
if not os.path.exists(os.path.join(odv_ref_png240, group, obj)):
os.mkdir(os.path.join(odv_ref_png240, group, obj))
for ref in os.listdir(os.path.join(org_dir, group, obj)):
vid_pth = os.path.join(org_dir, group, obj, ref)
ref_new = os.path.splitext(ref)[0]
if not os.path.exists(os.path.join(odv_ref_png240, group, obj, ref_new)):
os.mkdir(os.path.join(odv_ref_png240, group, obj, ref_new))
out_dir = os.path.join(odv_ref_png240, group, obj, ref_new, '%3d.png')
cmd = 'ffmpeg -f rawvideo -pixel_format yuv420p -s 480x240 -i {} {}'.format(vid_pth, out_dir)
os.system(cmd)
#
#
#
# for group in os.listdir(dataset_240):
# group_pth = os.path.join(dataset_240, group)
# for vid in os.listdir(group_pth):
# name, ext = os.path.splitext(vid)
# vid_pth = os.path.join(dataset_240, group, vid)
# if not os.path.exists(os.path.join(png240_dir, name)):
# os.mkdir(os.path.join(png240_dir, name))
# out_dir = os.path.join(png240_dir, name, '%3d.png')
# cmd = 'ffmpeg -f rawvideo -pixel_format yuv420p -s 480x240 -i {} {}'.format(vid_pth, out_dir)
# os.system(cmd)
# 将png240_dir中的图像 分配train到odv_train240_png
def shutil_train_test_png():
f = open(os.path.join(dataset_utils, 'train.txt'), "r")
lines = f.readlines() # 读取全部内容 ，并以列表方式返回
vid_all_list = []
keys_all_list = []
# num = 1
for num_ref in range(0, 540):
line = lines[num_ref]
# print(line)
line_list = line.split()
vid_id = line_list[0]
vid_id = int(vid_id)
# print(vid_id)
if vid_id in odv_train_id:
# print('!!!!!!!!!!')
# num+=1
# print('the train number is {}'.format(num))
vid_name = os.path.split(line_list[3])[-1].split('.yuv')[0]
# print(vid_name)
folder_path = os.path.join(odv_train240_png, vid_name)
# if not os.path.exists(folder_path):
# os.mkdir(folder_path)
shutil.copytree(os.path.join(png240_dir, vid_name), folder_path)
# 将png240_dir中的图像 分配test到odv_test240_png
def shutil_test_png():
f = open(os.path.join(dataset_utils, 'train.txt'), "r")
lines = f.readlines() # 读取全部内容 ，并以列表方式返回
vid_all_list = []
keys_all_list = []
# num = 1
for num_ref in range(0, 540):
line = lines[num_ref]
# print(line)
line_list = line.split()
vid_id = line_list[0]
vid_id = int(vid_id)
# print(vid_id)
if vid_id in odv_test_id:
# print('!!!!!!!!!!')
# num+=1
# print('the train number is {}'.format(num))
vid_name = os.path.split(line_list[3])[-1].split('.yuv')[0]
# print(vid_name)
folder_path = os.path.join(odv_test_png240, vid_name)
# if not os.path.exists(folder_path):
# os.mkdir(folder_path)
shutil.copytree(os.path.join(png240_dir, vid_name), folder_path)
# 2. 为训练BVQA360网络，制作train set的LMDB文件
def create_lmdb_for_odv_train480_new():
img_path_list, keys = prepare_keys_reds(odv_train240_png)
make_lmdb_from_imgs(odv_train240_png, odvLmdb_train_pth2, img_path_list, keys, multiprocessing_read=True)
def prepare_keys_reds(folder_path):
"""Prepare image path list and keys for REDS dataset.
Args:
folder_path (str): Folder path.
Returns:
list[str]: Image path list.
list[str]: Key list.
"""
print('Reading image path list ...')
img_path_list = sorted(
list(scandir(folder_path, suffix='png', recursive=True)))
keys = [v.split('.png')[0] for v in img_path_list] # example: 000/00000000
print(keys)
return img_path_list, keys
def make_lmdb_from_imgs(data_path,
lmdb_path,
img_path_list,
keys,
batch=5000,
compress_level=1,
multiprocessing_read=False,
n_thread=40,
map_size=None):
"""Make lmdb from images.
Contents of lmdb. The file structure is:
example.lmdb
├── data.mdb
├── lock.mdb
├── meta_info.txt
The data.mdb and lock.mdb are standard lmdb files and you can refer to
https://lmdb.readthedocs.io/en/release/ for more details.
The meta_info.txt is a specified txt file to record the meta information
of our datasets. It will be automatically created when preparing
datasets by our provided dataset tools.
Each line in the txt file records 1)image name (with extension),
2)image shape, and 3)compression level, separated by a white space.
For example, the meta information could be:
`000_00000000.png (720,1280,3) 1`, which means:
1) image name (with extension): 000_00000000.png;
2) image shape: (720,1280,3);
3) compression level: 1
We use the image name without extension as the lmdb key.
If `multiprocessing_read` is True, it will read all the images to memory
using multiprocessing. Thus, your server needs to have enough memory.
Args:
data_path (str): Data path for reading images.
lmdb_path (str): Lmdb save path.
img_path_list (str): Image path list.
keys (str): Used for lmdb keys.
batch (int): After processing batch images, lmdb commits.
Default: 5000.
compress_level (int): Compress level when encoding images. Default: 1.
multiprocessing_read (bool): Whether use multiprocessing to read all
the images to memory. Default: False.
n_thread (int): For multiprocessing.
map_size (int | None): Map size for lmdb env. If None, use the
estimated size from images. Default: None
"""
assert len(img_path_list) == len(keys), (
'img_path_list and keys should have the same length, '
f'but got {len(img_path_list)} and {len(keys)}')
print(f'Create lmdb for {data_path}, save to {lmdb_path}...')
print(f'Totoal images: {len(img_path_list)}')
if not lmdb_path.endswith('.lmdb'):
raise ValueError("lmdb_path must end with '.lmdb'.")
if osp.exists(lmdb_path):
print(f'Folder {lmdb_path} already exists. Exit.')
sys.exit(1)
if multiprocessing_read:
# read all the images to memory (multiprocessing)
dataset = {} # use dict to keep the order for multiprocessing
shapes = {}
print(f'Read images with multiprocessing, #thread: {n_thread} ...')
pbar = tqdm(total=len(img_path_list), unit='image')
def callback(arg):
"""get the image data and update pbar."""
key, dataset[key], shapes[key] = arg
pbar.update(1)
pbar.set_description(f'Read {key}')
pool = Pool(n_thread)
for path, key in zip(img_path_list, keys):
pool.apply_async(
read_img_worker,
args=(osp.join(data_path, path), key, compress_level),
callback=callback)
pool.close()
pool.join()
pbar.close()
print(f'Finish reading {len(img_path_list)} images.')
# create lmdb environment
if map_size is None:
# obtain data size for one image
img = cv2.imread(
osp.join(data_path, img_path_list[0]), cv2.IMREAD_UNCHANGED)
_, img_byte = cv2.imencode(
'.png', img, [cv2.IMWRITE_PNG_COMPRESSION, compress_level])
data_size_per_img = img_byte.nbytes
print('Data size per image is: ', data_size_per_img)
data_size = data_size_per_img * len(img_path_list)
map_size = data_size * 10
env = lmdb.open(lmdb_path, map_size=map_size)
# write data to lmdb
pbar = tqdm(total=len(img_path_list), unit='chunk')
txn = env.begin(write=True)
txt_file = open(osp.join(lmdb_path, 'meta_info.txt'), 'w')
for idx, (path, key) in enumerate(zip(img_path_list, keys)):
pbar.update(1)
pbar.set_description(f'Write {key}')
key_byte = key.encode('ascii')
if multiprocessing_read:
img_byte = dataset[key]
h, w, c = shapes[key]
else:
_, img_byte, img_shape = read_img_worker(
osp.join(data_path, path), key, compress_level)
h, w, c = img_shape
txn.put(key_byte, img_byte)
# write meta information
txt_file.write(f'{key}.png ({h},{w},{c}) {compress_level}\n')
if idx % batch == 0:
txn.commit()
txn = env.begin(write=True)
pbar.close()
txn.commit()
env.close()
txt_file.close()
print('\nFinish writing lmdb.')
def read_img_worker(path, key, compress_level):
"""Read image worker.
Args:
path (str): Image path.
key (str): Image key.
compress_level (int): Compress level when encoding images.
Returns:
str: Image key.
byte: Image byte.
tuple[int]: Image shape.
"""
img = cv2.imread(path, cv2.IMREAD_UNCHANGED)
if img.ndim == 2:
h, w = img.shape
c = 1
else:
h, w, c = img.shape
_, img_byte = cv2.imencode('.png', img,
[cv2.IMWRITE_PNG_COMPRESSION, compress_level])
return (key, img_byte, (h, w, c))
class LmdbMaker():
"""LMDB Maker.
Args:
lmdb_path (str): Lmdb save path.
map_size (int): Map size for lmdb env. Default: 1024 ** 4, 1TB.
batch (int): After processing batch images, lmdb commits.
Default: 5000.
compress_level (int): Compress level when encoding images. Default: 1.
"""
def __init__(self,
lmdb_path,
map_size=1024**4,
batch=5000,
compress_level=1):
if not lmdb_path.endswith('.lmdb'):
raise ValueError("lmdb_path must end with '.lmdb'.")
if osp.exists(lmdb_path):
print(f'Folder {lmdb_path} already exists. Exit.')
sys.exit(1)
self.lmdb_path = lmdb_path
self.batch = batch
self.compress_level = compress_level
self.env = lmdb.open(lmdb_path, map_size=map_size)
self.txn = self.env.begin(write=True)
self.txt_file = open(osp.join(lmdb_path, 'meta_info.txt'), 'w')
self.counter = 0
def put(self, img_byte, key, img_shape):
self.counter += 1
key_byte = key.encode('ascii')
self.txn.put(key_byte, img_byte)
# write meta information
h, w, c = img_shape
self.txt_file.write(f'{key}.png ({h},{w},{c}) {self.compress_level}\n')
if self.counter % self.batch == 0:
self.txn.commit()
self.txn = self.env.begin(write=True)
def close(self):
self.txn.commit()
self.env.close()
self.txt_file.close()
# 本项目test时使用PNG图像输入，这个函数是用于制作所需的所有test PNG图像到一个目录下
def odv_test_png240_ForTest():
f1 = open(os.path.join('/home/yl/bvqa360/data/', 'test_240ODV.txt'), "r")
lines = f1.readlines()
for num_ref in range(0, 108):
line = lines[num_ref]
# print(line)
line_list = line.split()
vid_id = line_list[1]
print(vid_id)
for im in os.listdir(os.path.join(png240_dir, vid_id)):
print(im)
tar = os.path.join(new_data_png240, vid_id) + '_' + im
print(tar)
src = os.path.join(png240_dir, vid_id, im)
shutil.copyfile(src, tar)
#########################################################################################################################################################
##### Prepare any kinds of imformation and configurarion for the IQA/VQA/BVQA compared algrorithm #####
#########################################################################################################################################################
################################################
##### Dataset path for compared algorithms #####
################################################
saved_log = r'/home/yl/PSNR_BVQA360/'
# ./360tools_metric
# < Usage >
# -c, --config [STRING] (optional)
# : config file name
# -o, --original [STRING]
# : original file name
# -r, --reconstruct [STRING]
# : output file name
# -w, --original_width [INTEGER]
# : width of original image
# -h, --original_height [INTEGER]
# : height of original image
# -l, --reconstructed_width [INTEGER] (optional)
# : width of reconstructed image
# -m, --reconstructed_height [INTEGER] (optional)
# : height of reconstruncted image
# -n, --num_frame [INTEGER] (optional)
# : number of frames to be processed
# -q, --qmetric [INTEGER]
# : quality metric
# 1: PSNR
# 2: sperical PSNR(S-PSNR)
# 3: weighted sperical PSNR(WS-PSNR)
# 4: craster parabolic projection PSNR(CPP-PSNR)
# -s, --sphere_file [STRING] (optional)
# : sphere file; used for calculation of S-PSNR
# -f, --original_proj_format [INTEGER] (optional)
# : original projection format
# 1: ERP
# 2: ISP
# 3: CMP
# 4: OHP
# 5: TSP
# 6: CPP
# 7: SSP
# -t, --reconstructed_proj_format [INTEGER] (optional)
# : reconstructed projection format
# 1: ERP
# 2: ISP
# 3: CMP
# 4: OHP
# 5: TSP
# 6: CPP
# 7: SSP
# -x, --color_space_orig [INTEGER] (optional)
# : color space
# 1: YUV420 8-bit
# 2: YUV420 10-bit
#
# -y, --color_space_ref [INTEGER] (optional)
# : color space
# 1: YUV420 8-bit
# 2: YUV420 10-bit
#
# -v, --verbose [FLAG] (optional)
# : verbose output
### 1. PSNR/S-PSNE/WS-PSNR/CPP-PSNR
# 执行：放在360tools/bin里面
def countPSNR():
# spsnrtxt = '/home/liutie/Desktop/360tools-master/bin/sphere655362.txt'
log_pth = os.path.join(saved_log, 'psnr_spsnr++_dmos.txt')
f = open(os.path.join(dataset_utils, 'train.txt'), "r")
lines = f.readlines() # 读取全部内容 ，并以列表方式返回
namelist = []
i = 0
for num_ref in range(0, 540):
line = lines[num_ref]
line_list = line.split() # string to list
vid_id = line_list[0]
vid_id = int(vid_id)
if vid_id in odv_test_id:
# print('!!!!!!!!!!')
# num+=1
# print('the train number is {}'.format(num))
imp_name = os.path.split(line_list[3])[-1]
ref_name = os.path.split(line_list[2])[-1]
group_name = 'Group' + imp_name[1]
if num_ref >= (540 - 54):
group_name = 'Group' + str(10)
ref_pth = os.path.join(orig_data, group_name, 'Reference', ref_name)
test_pth = os.path.join(orig_data, group_name, imp_name)
w = int(line_list[5])
h = int(line_list[6])
num_frame = int(line_list[7])
i += 1
print('this is {}-th test odv ##############'.format(i))
for metric in range(1, 5):
if metric == 1:
cmd1 = r'./360tools_metric -w {} -h {} -n {} -f 1 -o {} -q {} -l {} -m {} -t 1 -r {} >> {}' # -n: frames
cmd1 = cmd1.format(w, h, num_frame, ref_pth, str(metric), w, h, test_pth, log_pth)
os.system(cmd1)
print(cmd1)
elif metric == 2:
cmd2 = r'./360tools_metric -w {} -h {} -n {} -f 1 -o {} -q {} -l {} -m {} -s sphere_655362.txt -t 1 -r {} >> {}'
cmd2 = cmd2.format(w, h, num_frame, ref_pth, str(metric), w, h, test_pth, log_pth)
os.system(cmd2)
print(cmd2)
elif metric == 3:
cmd3 = r'./360tools_metric -w {} -h {} -n {} -f 1 -o {} -q {} -l {} -m {} -t 1 -r {} >> {}' # -n: frames
cmd3 = cmd3.format(w, h, num_frame, ref_pth, str(metric), w, h, test_pth, log_pth)
os.system(cmd3)
print(cmd3)
elif metric == 4:
cmd4 = r'./360tools_metric -w {} -h {} -n {} -f 1 -o {} -q {} -l {} -m {} -t 1 -r {} >> {}'
cmd4 = cmd4.format(w, h, num_frame, ref_pth, str(metric), w, h, test_pth, log_pth)
os.system(cmd4)
print(cmd4)
# extract the Y channel psnr.. shape:(960, 4)
def cal_Y_PSNR():
psnr = np.loadtxt(os.path.join(saved_log, 'psnr_spsnr++_dmos.txt'), dtype=np.str)[:, 1].astype(np.float32).reshape((108, 4))
#print(psnr.dtype)
np.savetxt(os.path.join(saved_log, 'PSNR_S-PSNE_WS-PSNR_CPP-PSNR_dmos.txt') , psnr, fmt='%.04f')
### 1. PSNR/S-PSNE/WS-PSNR/CPP-PSNR
# 执行：放在360tools/bin里面
def countPSNR_BIT():
# spsnrtxt = '/home/liutie/Desktop/360tools-master/bin/sphere655362.txt'
log_pth = os.path.join(saved_log, 'psnr_spsnr++_BIT_dmos4.txt')
f = open(r'/home/yl/bvqa360/bit_dataset/score.txt', "r")
lines = f.readlines() # 读取全部内容 ，并以列表方式返回
vid_dmos = []
i = 0
for idx in range(len(lines)):
line = lines[idx]
imp_name = line.split()[0]
id = imp_name.split('-')[0]
ref_vid = id + '-' + '100MB' + '.yuv'
imp_vid = imp_name + '.yuv'
ref_pth = os.path.join(data_BIT_ref_yuv240, ref_vid)
test_pth = os.path.join(data_BIT_imp_yuv240, imp_vid)
w = 480
h = 240
num_frame = 10
i += 1
print('this is {}-th test odv ##############'.format(i))
for metric in range(1, 5):
if metric == 1:
cmd1 = r'./360tools_metric -w {} -h {} -n {} -f 1 -o {} -q {} -l {} -m {} -t 1 -r {} >> {}' # -n: frames
cmd1 = cmd1.format(w, h, num_frame, ref_pth, str(metric), w, h, test_pth, log_pth)
os.system(cmd1)
print(cmd1)
elif metric == 2:
cmd2 = r'./360tools_metric -w {} -h {} -n {} -f 1 -o {} -q {} -l {} -m {} -s sphere_655362.txt -t 1 -r {} >> {}'
cmd2 = cmd2.format(w, h, num_frame, ref_pth, str(metric), w, h, test_pth, log_pth)
os.system(cmd2)
print(cmd2)
elif metric == 3:
cmd3 = r'./360tools_metric -w {} -h {} -n {} -f 1 -o {} -q {} -l {} -m {} -t 1 -r {} >> {}' # -n: frames
cmd3 = cmd3.format(w, h, num_frame, ref_pth, str(metric), w, h, test_pth, log_pth)
os.system(cmd3)
print(cmd3)
elif metric == 4:
cmd4 = r'./360tools_metric -w {} -h {} -n {} -f 1 -o {} -q {} -l {} -m {} -t 1 -r {} >> {}'
cmd4 = cmd4.format(w, h, num_frame, ref_pth, str(metric), w, h, test_pth, log_pth)
os.system(cmd4)
print(cmd4)
# extract the Y channel psnr.. shape:(960, 4)
def cal_Y_PSNR_BIT():
psnr = np.loadtxt(os.path.join(saved_log, 'psnr_spsnr++_BIT_dmos4.txt'), dtype=np.str)[:, 1].astype(np.float32).reshape((144, 4))
#print(psnr.dtype)
np.savetxt(os.path.join(saved_log, 'BIT_PSNR_S-PSNE_WS-PSNR_CPP-PSNR_dmos4.txt') , psnr, fmt='%.04f')
# odv_test_png240 = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_yl240_png_test' # test 测试集 108个目录的png
# odv_ref_png240 = r'/media/yl/yl_8t/data_bvqa_lc_resize/data_vqa_yl240_png_ref'
def cal_WSSSIM():
log_pth = os.path.join(saved_log, 'WS-SSIM_dmos.txt')
f = open(os.path.join(dataset_utils, 'train.txt'), "r")
lines = f.readlines() # 读取全部内容 ，并以列表方式返回
vid_dmos = []
i = 0
for num_ref in range(0, 540):
line = lines[num_ref]
line_list = line.split() # string to list
vid_id = line_list[0]
vid_id = int(vid_id)
if vid_id in odv_test_id:
# print('!!!!!!!!!!')
# num+=1
# print('the train number is {}'.format(num))
imp_name = os.path.split(line_list[3])[-1]
ref_name = os.path.split(line_list[2])[-1]
print(imp_name)
num_frame = int(line_list[7])
group_name = 'Group' + imp_name[1]
if num_ref >= (540 - 54):
group_name = 'Group' + str(10)
imp_name = os.path.splitext(imp_name)[0]
ref_name = os.path.splitext(ref_name)[0]
ref_pth = os.path.join(odv_ref_png240, group_name, 'Reference', ref_name)
print(ref_pth)
test_pth = os.path.join(odv_test_png240, imp_name)
frame_dmos = 0
for test_frame in os.listdir(test_pth):
test_frame_pth = os.path.join(test_pth, test_frame)
ref_frame_pth = os.path.join(ref_pth, test_frame)
test = cv2.imread(test_frame_pth, cv2.IMREAD_COLOR)
test = cv2.cvtColor(test, cv2.COLOR_BGR2RGB).astype(np.uint8)
#print(test.shape)
ref = cv2.imread(ref_frame_pth, cv2.IMREAD_COLOR)
#print(ref)
ref = cv2.cvtColor(ref, cv2.COLOR_BGR2RGB).astype(np.uint8)
test = cv2.cvtColor(test, cv2.COLOR_RGB2GRAY)
ref= cv2.cvtColor(ref, cv2.COLOR_RGB2GRAY)
# ref = np.array(Image.open(ref_frame_pth))
# test = np.array(Image.open(test_frame_pth))
frame_dmos = frame_dmos + ws_ssim(ref, test) / num_frame
vid_dmos.append(frame_dmos)
np.savetxt(log_pth, np.array(vid_dmos))
def cal_BIT_WSSSIM():
log_pth = os.path.join(saved_log, 'WS-SSIM_BIT_dmos5.txt')
f = open(r'/home/yl/bvqa360/bit_dataset/score.txt', "r")
lines = f.readlines() # 读取全部内容 ，并以列表方式返回
vid_dmos = []
i = 0
interval = 300
num_frame = 300 / interval
for idx in range(len(lines)):
line = lines[idx]
imp_name = line.split()[0]
id = imp_name.split('-')[0]
ref_vid = id + '-' + '100MB'
imp_vid = imp_name
ref_pth = os.path.join(data_BIT_ref_png240, ref_vid)
# print(ref_pth)
test_pth = os.path.join(data_BIT_imp_png240, imp_vid)
frame_dmos = 0
for j in range(1, 301, interval):
test_frame = '%03d'%j + '.png'
#for test_frame in os.listdir(test_pth):
test_frame_pth = os.path.join(test_pth, test_frame)
ref_frame_pth = os.path.join(ref_pth, test_frame)
test = cv2.imread(test_frame_pth, cv2.IMREAD_COLOR)
test = cv2.cvtColor(test, cv2.COLOR_BGR2RGB).astype(np.uint8)
# print(test.shape)
ref = cv2.imread(ref_frame_pth, cv2.IMREAD_COLOR)
# print(ref)
ref = cv2.cvtColor(ref, cv2.COLOR_BGR2RGB).astype(np.uint8)
test = cv2.cvtColor(test, cv2.COLOR_RGB2GRAY)
ref = cv2.cvtColor(ref, cv2.COLOR_RGB2GRAY)
# ref = np.array(Image.open(ref_frame_pth))
# test = np.array(Image.open(test_frame_pth))
frame_dmos = frame_dmos + ws_ssim(ref, test) / num_frame
vid_dmos.append(frame_dmos)
np.savetxt(log_pth, np.array(vid_dmos))
#########################################################################################################################################################
##### Prepare the dataset of BIT360video dataset[1] for validate the generalization ability of the IQA/VQA/BVQA compared algrorithms #####
#########################################################################################################################################################
# [1] SUBJECTIVE AND OBJECTIVE QUALITY ASSESSMENT OF PANORAMIC VIDEOS IN VIRTUAL REALITY ENVIRONMENTS,