/
rafdb_dataset_hq.py
164 lines (138 loc) · 6.35 KB
/
rafdb_dataset_hq.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
import numpy as np
import time
import random
import cv2
import sys
import os
from dataset_tools import enclosing_square, add_margin, cut
import keras
from tqdm import tqdm
from dataset_tools import _readcsv, cntk_filtering
from dataset_tools import linear_balance_illumination, mean_std_normalize, equalize_hist
from dataset_tools import draw_emotion
from dataset_tools import DataGenerator
from six.moves import cPickle as pickle
NUM_TRAINING_SAMPLES = 12271
NUM_CLASSES = 7
def _crop_align_rafdb_face(im, ann, TARGET_SIZE = (256,256)):
p = []
for i in range(5):
p.append( [float(x) for x in ann[i].strip().replace(' ','\t').split('\t')] )
pmouth=( (p[4][0]+p[3][0])//2 , (p[4][1]+p[3][1])//2 )
srcTri = np.array( [ p[0], p[1], pmouth ] ).astype(np.float32)
dstTri = np.array( [ [.286, .4], [.714, .4], [.50, .714] ] ).astype(np.float32) * (np.array(TARGET_SIZE, dtype=np.float32).reshape((1,2)))
dh = TARGET_SIZE[0] - im.shape[0]
dw = TARGET_SIZE[1] - im.shape[1]
if dh > 0:
pad = np.zeros((dh, im.shape[1], im.shape[2]), dtype=np.uint8)
im = np.append(im, pad, axis=0)
if dw > 0:
pad = np.zeros((im.shape[0], dw, im.shape[2]), dtype=np.uint8)
im = np.append(im, pad, axis=1)
A = cv2.getAffineTransform(srcTri,dstTri)
im = cv2.warpAffine(im, A, (im.shape[1], im.shape[0]))
return im[0:TARGET_SIZE[0],0:TARGET_SIZE[1],:]
def _load_rafdb(meta, imagesdir, partition):
data = []
n_discarded=0
n_discarded_cntk=0
for n,d in enumerate(tqdm(meta)):
actualpartition = 'train' if n<NUM_TRAINING_SAMPLES else 'test'
if actualpartition == partition:
imgid = '%05d'%(n+1) if n<NUM_TRAINING_SAMPLES else '%04d'%(n+1-NUM_TRAINING_SAMPLES)
drop, labels = cntk_filtering(d, rowtotal=1, num_classes=NUM_CLASSES)
if not drop:
path = os.path.join(imagesdir,'%s_%s.jpg'%(partition,imgid))
annpath= os.path.join(imagesdir, '..', '..', 'Annotation', 'manual','%s_%s_manu_attri.txt'%(partition,imgid))
ann = [l for l in open(annpath)]
img = cv2.imread(path)
img = _crop_align_rafdb_face(img, ann)
if img is not None:
example={
'img': img,
'label': labels,
'roi': (16,16,224,224)
}
if np.max(example['img'])==np.min(example['img']):
print('Warning, blank image!')
else:
data.append(example)
else: # img is None
print("WARNING! Unable to read %s" % path)
n_discarded+=1
else: # ambiguous label
n_discarded_cntk+=1
print("Data loaded. %d samples (%d+%d discarded)" % (len(data), n_discarded, n_discarded_cntk) )
return data
class RAFDBDataset:
def __init__(self, partition='train', imagesdir='RAF-DB/basic/Image/original', csvmeta='RAF-DB/distribute_basic.csv', target_shape=(224,224,3), augment=True, custom_augmentation=None, preprocessing='full_normalization', debug_max_num_samples=None):
if partition.startswith('train'):
partition='train'
elif partition.startswith('val'):
print('WARNING: this dataset only has one test partition for test and validation')
partition='test'
elif partition.startswith('test'):
print('WARNING: this dataset only has one test partition for test and validation')
partition='test'
else:
raise Exception("unknown partition")
self.target_shape = target_shape
self.custom_augmentation = custom_augmentation
self.augment = augment
self.gen = None
self.preprocessing = preprocessing
print('Loading data...')
cache_file_name = 'hq%s.%s%s.cache'%(imagesdir.replace('/','_'),partition, '.'+str(debug_max_num_samples) if debug_max_num_samples is not None else '')
try:
with open(cache_file_name, 'rb') as f:
self.data = pickle.load(f)
self.data = self.data[:debug_max_num_samples]
print("Data loaded. %d samples, from cache" % (len(self.data)) )
except FileNotFoundError:
meta = _readcsv(csvmeta)
print('csv read complete: %d.' %(len(meta)))
self.data = _load_rafdb(meta, imagesdir, partition)
with open(cache_file_name, 'wb') as f:
pickle.dump(self.data, f)
def get_num_samples(self):
return self.data.shape[0]
def get_num_classes(self):
return NUM_CLASSES
def get_generator(self, batch_size=64):
if self.gen is None:
self.gen = DataGenerator(self.data, self.target_shape, with_augmentation=self.augment, custom_augmentation=self.custom_augmentation, batch_size=batch_size, preprocessing=self.preprocessing)
return self.gen
def test1():
print('Training')
dt = RAFDBDataset(target_shape=(224,224,3), preprocessing='full_normalization', debug_max_num_samples=None, augment=False)
print('Test')
dv = RAFDBDataset('test',target_shape=(200,200,3), preprocessing='full_normalization', debug_max_num_samples=None)
print('Now generating from training set')
gen = dt.get_generator()
i=0
while True:
print(i)
i+=1
for batch in tqdm(gen):
pass
EMOTION_LABELS=['surprise','fear','disgust','happiness','sadness','anger','neutral']
def test2():
dt = RAFDBDataset(target_shape=(200,200,3), preprocessing='full_normalization', debug_max_num_samples=None, augment=False)
gen = dt.get_generator()
for batch in gen:
for x,y in zip(batch[0], batch[1]):
window = np.zeros((400,200,3), dtype=np.uint8)
MAX = np.amax(x)
MIN = np.amin(x)
x = 255*(x-MIN)/(MAX-MIN)
x = x.clip(0,255).astype(np.uint8)
if len(x.shape)<=2 or x.shape[2]==1:
x = cv2.cvtColor(x, cv2.COLOR_GRAY2BGR)
window[0:200,0:200,:] = x
window[200:400,0:200,:] = draw_emotion(y,200,200, emotion_labels=EMOTION_LABELS)
cv2.imshow('im', window)
k = cv2.waitKey(0)
if k==27:
sys.exit(0)
if '__main__' == __name__:
test2()