forked from Netflix/image_compression_comparison
-
Notifications
You must be signed in to change notification settings - Fork 0
/
compute_BD_rates.py
176 lines (146 loc) · 8.12 KB
/
compute_BD_rates.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
import sqlite3
import sys
import logging
import ntpath
from statistics import mean
from collections import namedtuple
from collections import defaultdict
from bd_rate_calculator import BDrateCalculator
from analyze_encoding_results import apply_size_check
__author__ = "Aditya Mavlankar"
__copyright__ = "Copyright 2019-2020, Netflix, Inc."
__credits__ = ["Kyle Swanson", "Jan de Cock", "Marjan Parsa"]
__license__ = "Apache License, Version 2.0"
__version__ = "0.1"
__maintainer__ = "Aditya Mavlankar"
__email__ = "[email protected]"
__status__ = "Development"
RateQualityPoint = namedtuple('RateQualityPoint', ['bpp', 'quality', 'target_metric', 'target_value'])
BD_RATE_EXCEPTION_STRING = 'BD_RATE_EXCEPTION'
def get_unique_sources_sorted(connection):
unique_sources = connection.execute('SELECT DISTINCT SOURCE FROM ENCODES').fetchall()
unique_sources = [elem[0] for elem in unique_sources]
return sorted(list(set(unique_sources)))
def get_quality_dict(elem, list_of_metrics):
quality = dict()
for index, metric in enumerate(list_of_metrics):
quality[metric] = elem[index]
return quality
def get_rate_quality_points(connection, sub_sampling, codec, source, total_pixels, list_of_metrics):
# print('{} {} {}'.format(codec, sub_sampling, source))
csv_metrics_upper = ','.join([elem.upper() for elem in list_of_metrics])
points = connection.execute("SELECT {},FILE_SIZE_BYTES,TARGET_METRIC,TARGET_VALUE FROM ENCODES WHERE CODEC='{}' AND SUB_SAMPLING='{}' AND SOURCE='{}'"
.format(csv_metrics_upper, codec, sub_sampling, source)).fetchall()
rate_quality_points = [
RateQualityPoint(elem[len(list_of_metrics)] * 8 / total_pixels, get_quality_dict(elem, list_of_metrics),
elem[len(list_of_metrics) + 1], elem[len(list_of_metrics) + 2]) for elem in points]
# print(repr(rate_quality_points))
return rate_quality_points
def get_rates(rate_quality_points):
return [rate_quality_point.bpp for rate_quality_point in rate_quality_points]
def get_quality(rate_quality_points, metric):
return [rate_quality_point.quality[metric] for rate_quality_point in rate_quality_points]
def get_formatted_bdrate(val):
if isinstance(val, str):
return val
else:
return '{:.2f}'.format(val).rjust(6)
def get_formatted_mean_bdrate(val):
return '{:.2f}'.format(val).rjust(22)
def my_shorten(name, width):
return (name[:width - 3] + '...') if len(name) > width else name
def print_bd_rates(bdrates_various_metrics, codec, unique_sources, black_list_source_various_metrics,
list_of_metrics):
bdrates_this_codec_various_metrics = dict()
for metric in list_of_metrics:
bdrates_this_codec_various_metrics[metric] = list()
max_len_source_name = len(max(unique_sources, key=len))
max_len_to_use_for_printing = min(80, max_len_source_name)
for source in unique_sources:
print_string = ' {} {}'.format(my_shorten(source,
max_len_to_use_for_printing)
.ljust(max_len_to_use_for_printing),
codec)
for metric in list_of_metrics:
print_string += ' BDRate-{} {}'.format(metric.upper(), get_formatted_bdrate(bdrates_various_metrics[metric][codec][source]))
if source not in black_list_source_various_metrics[metric]:
bdrates_this_codec_various_metrics[metric].append(bdrates_various_metrics[metric][codec][source])
print(print_string)
result = codec.ljust(16)
result_local = result
for metric in list_of_metrics:
result += '{}'.format(get_formatted_mean_bdrate(mean(bdrates_this_codec_various_metrics[metric])))
result_local += ' Mean BDRate-{} {:.2f}'.format(metric.upper(), mean(bdrates_this_codec_various_metrics[metric]))
print(result_local + '\n')
return result
def main(argv):
db_file_name = 'encoding_results_vmaf.db'
if len(argv) > 0:
db_file_name = argv[0]
connection = sqlite3.connect(db_file_name)
logger = logging.getLogger('report.bdrates')
logger.addHandler(logging.FileHandler('bdrates_' + ntpath.basename(db_file_name) + '.txt'))
logger.addHandler(logging.StreamHandler())
logger.setLevel('DEBUG')
unique_sources = get_unique_sources_sorted(connection)
total_pixels = apply_size_check(connection)
baseline_codec = 'jpeg'
sub_sampling_arr = ['420', '444']
codecs = ['jpeg-mse', 'jpeg-ms-ssim', 'jpeg-im', 'jpeg-hvs-psnr', 'webp', 'kakadu-mse', 'kakadu-visual', 'openjpeg',
'hevc', 'avif-mse', 'avif-ssim',
'avifenc-sp-0', 'avifenc-sp-2', 'avifenc-sp-4', 'avifenc-sp-6', 'avifenc-sp-8',
'avifenc-sp-0-crf', 'avifenc-sp-2-crf', 'avifenc-sp-4-crf', 'avifenc-sp-6-crf', 'avifenc-sp-8-crf']
metrics_for_BDRate = ['vmaf', 'ssim', 'ms_ssim', 'vif', 'adm2', 'psnr_y', 'psnr_avg']
for sub_sampling in sub_sampling_arr:
bdrates_various_metrics = dict()
black_list_source_various_metrics = dict()
for metric in metrics_for_BDRate:
bdrates_various_metrics[metric] = defaultdict(dict)
black_list_source_various_metrics[metric] = list()
print('\n\nComputing BD rates for subsampling {}'.format(sub_sampling))
for source in unique_sources:
baseline_rate_quality_points = get_rate_quality_points(connection, sub_sampling, baseline_codec, source, total_pixels, metrics_for_BDRate)
for codec in codecs:
if codec == 'webp' and sub_sampling == '444':
continue
rate_quality_points = get_rate_quality_points(connection, sub_sampling, codec, source, total_pixels, metrics_for_BDRate)
for metric in metrics_for_BDRate:
# print(metric.upper())
try:
bd_rate_val = 100.0 * BDrateCalculator.CalcBDRate(
list(zip(get_rates(baseline_rate_quality_points), get_quality(baseline_rate_quality_points, metric))),
list(zip(get_rates(rate_quality_points), get_quality(rate_quality_points, metric))))
bdrates_various_metrics[metric][codec][source] = bd_rate_val
except AssertionError as e:
print('{} {} {} {}: '.format(metric, source, codec, sub_sampling) + str(e))
bdrates_various_metrics[metric][codec][source] = BD_RATE_EXCEPTION_STRING
# BD rate computation failed for one of the codecs,
# so to be fair, ignore this source for final results
if source not in black_list_source_various_metrics[metric]:
black_list_source_various_metrics[metric].append(source)
for metric in metrics_for_BDRate:
print('{} black list {} BD RATE\n '.format(sub_sampling, metric.upper()) + repr(black_list_source_various_metrics[metric]))
results = dict()
for codec in codecs:
if codec == 'webp' and sub_sampling == '444':
continue
print('Codec {} subsampling {}, BD rates:'.format(codec, sub_sampling))
result = print_bd_rates(bdrates_various_metrics, codec, unique_sources,
black_list_source_various_metrics,
metrics_for_BDRate)
results[codec] = result
logger.info('\n\n===================' + '=' * 22 * len(metrics_for_BDRate))
results_header = 'Results for subsampling {}'.format(sub_sampling)
logger.info(results_header)
logger.info('-' * len(results_header))
table_header = 'Codec'.ljust(16)
for metric in metrics_for_BDRate:
table_header += ' Mean BDRate-{}'.format(metric.upper()).rjust(22)
logger.info(table_header)
for codec in codecs:
if codec == 'webp' and sub_sampling == '444':
continue
logger.info(results[codec])
logger.info('===================' + '=' * 22 * len(metrics_for_BDRate))
if __name__ == '__main__':
main(sys.argv[1:])