def generateSamples(trainData, data):
t = 0
print '##########################################################################'
for (imgPath, landmarkGt, bbox) in data:
img = cv2.imread(imgPath, cv2.CV_LOAD_IMAGE_GRAYSCALE)
assert (img is not None)
logger("process %s" % imgPath)
height, width = img.shape[:2]
#downsampled by 3: 3x3 patch
cephaImg = cv2.resize(img, (int(width / 3), int(height / 3)),
interpolation=cv2.INTER_NEAREST)
#raw data
#trainData,t = getData(trainData,landmarkGt,cephaImg,t)
#print ('After getting raw data,there are %d datas') % t
r1 = 20 / 3
r2 = 20 #60/3
r3 = 400 #400/3
for idx, landmark in enumerate(landmarkGt): #19个landmark
# 25 Positive samples
landmarkPs25 = randomShiftWithArgument(landmark, 0, r1, 25)
trainData, t = getData(trainData, landmarkPs25, cephaImg, t)
print('After getting 25 positive samples,there are %d datas') % t
# 500 negative samples
landmarkNs500 = randomShiftWithArgument(landmark, r2, r3, 500)
trainData, t = getData(trainData, landmarkNs500, cephaImg, t)
print(
'After getting 25 positive and 500 negative samples,there are %d datas'
) % t
if idx == 1:
break
return trainData
def generateSamples(trainData, data, landmarks):
t = 0
for (imgPath, bbox) in data:
img = cv2.imread(imgPath, cv2.CV_LOAD_IMAGE_GRAYSCALE)
assert (img is not None)
logger('process %s' % imgPath)
height, width = img.shape[:2]
#downsampled by 3: 3X3 Patch
size = (int(width / 3), int(height / 3))
img = cv2.resize(img, size, interpolation=cv2.INTER_NEAREST)
trainData, t = getTrainData(trainData, landmarks, img)
print('After getting raw data,there are %d datas') % t
r2 = 20 / 3
r3 = 10 #400 / 3
for idx, landmark in enumerate(landmarks):
print '@@@@@@@@@@@@@@' + str(idx)
# 25 Positive samples
landmarkPs25 = randomShiftWithArgument(landmark, 0, r2, 25)
trainData, t = getTrainData(trainData, landmarkPs25, img)
#print ('After getting 25 positive samples,there are %d datas') % t
# 500 negative samples
landmarkNs500 = randomShiftWithArgument(landmark, r2, r3, 500)
trainData, t = getTrainData(trainData, landmarkNs500, img)
print(
'After getting 25 positive and 500 negative samples,there are %d datas'
) % t
return trainData
def generate(ftxt, mode, argument=False):
'''
第二阶段数据源制作
:param ftxt: 数据源文件位置和label
:param mode: 训练或测试
:param argument:
:return:
'''
data = getDataFromTxt(ftxt)
trainData = defaultdict(lambda: dict(patches=[], landmarks=[]))
for (imgPath, bbox, landmarkGt) in data:
img = cv2.imread(imgPath, cv2.IMREAD_GRAYSCALE)
assert (img is not None)
logger("process %s" % imgPath)
landmarkPs = randomShiftWithArgument(landmarkGt, 0.05)
if not argument:
landmarkPs = [landmarkPs[0]]
for landmarkP in landmarkPs:
for idx, name, padding in types:
patch, patch_bbox = getPatch(img, bbox, landmarkP[idx],
padding)
patch = cv2.resize(patch, (15, 15))
patch = patch.reshape((1, 15, 15))
trainData[name]['patches'].append(patch)
_ = patch_bbox.project(bbox.reproject(landmarkGt[idx]))
trainData[name]['landmarks'].append(_)
for idx, name, padding in types:
logger('writing training data of %s' % name)
patches = np.asarray(trainData[name]['patches'])
landmarks = np.asarray(trainData[name]['landmarks'])
patches = processImage(patches)
shuffle_in_unison_scary(patches, landmarks)
with h5py.File(
'/python/face_key_point/data_hdf5/train/2_%s/%s.h5' %
(name, mode), 'w') as h5:
h5['data'] = patches.astype(np.float32)
h5['landmark'] = landmarks.astype(np.float32)
with open(
'/python/face_key_point/data_hdf5/train/2_%s/%s.txt' %
(name, mode), 'w') as fd:
fd.write('/python/face_key_point/data_hdf5/train/2_%s/%s.h5' %
(name, mode))
def generate(ftxt, mode, argument=False):
"""
Generate Training Data for LEVEL-3
mode = train or test
"""
data = getDataFromTxt(ftxt)
trainData = defaultdict(lambda: dict(patches=[], landmarks=[]))
for (imgPath, bbox, landmarkGt) in data:
img = cv2.imread(imgPath, cv2.CV_LOAD_IMAGE_GRAYSCALE)
assert (img is not None)
logger("process %s" % imgPath)
landmarkPs = randomShiftWithArgument(landmarkGt, 0.01)
if not argument:
landmarkPs = [landmarkPs[0]]
for landmarkP in landmarkPs:
for idx, name, padding in types:
patch, patch_bbox = getPatch(img, bbox, landmarkP[idx],
padding)
patch = cv2.resize(patch, (15, 15))
patch = patch.reshape((1, 15, 15))
trainData[name]['patches'].append(patch)
_ = patch_bbox.project(bbox.reproject(landmarkGt[idx]))
trainData[name]['landmarks'].append(_)
for idx, name, padding in types:
logger('writing training data of %s' % name)
patches = np.asarray(trainData[name]['patches'])
landmarks = np.asarray(trainData[name]['landmarks'])
patches = processImage(patches)
shuffle_in_unison_scary(patches, landmarks)
with h5py.File(
'/home/tyd/下载/deep_landmark/mydataset/mytrain/3_%s/%s.h5' %
(name, mode), 'w') as h5:
h5['data'] = patches.astype(np.float32)
h5['landmark'] = landmarks.astype(np.float32)
with open(
'/home/tyd/下载/deep_landmark/mydataset/mytrain/3_%s/%s.txt' %
(name, mode), 'w') as fd:
fd.write(
'/home/tyd/下载/deep_landmark/mydataset/mytrain/3_%s/%s.h5' %
(name, mode))
def generateSamples(testData,data,landmarks):
for (imgPath,bbox) in data:
img = cv2.imread(imgPath,cv2.CV_LOAD_IMAGE_GRAYSCALE)
assert(img is not None)
logger('process %s' % imgPath)
height,width = img.shape[:2]
#downsampled by 3: 3X3 Patch
#size = (int(width/3),int(height/3))
size = (width,height)#不进行下采样
img = cv2.resize(img,size,interpolation=cv2.INTER_NEAREST)
testData = getData(testData,landmarks,img)#test此时为一个字典，存储样本点和对应的图像块
print ('After getting raw data,there are %d datas') % len(testData)
for idx,landmark in enumerate(landmarks):
# 产生样本点 samples
landmark_samples = randomShiftWithArgument(landmark,0,100,150)
testData = getData(testData,landmark_samples,img)
return testData
def generate(ftxt, mode, argument=False):
"""
Generate Training Data for LEVEL-3
mode = train or test
"""
data = getDataFromTxt(ftxt)
trainData = defaultdict(lambda: dict(patches=[], landmarks=[]))
for (imgPath, bbox, landmarkGt) in data:
img = cv2.imread(imgPath, cv2.CV_LOAD_IMAGE_GRAYSCALE)
assert(img is not None)
logger("process %s" % imgPath)
landmarkPs = randomShiftWithArgument(landmarkGt, 0.01)
if not argument:
landmarkPs = [landmarkPs[0]]
for landmarkP in landmarkPs:
for idx, name, padding in types:
patch, patch_bbox = getPatch(img, bbox, landmarkP[idx], padding)
patch = cv2.resize(patch, (15, 15))
patch = patch.reshape((1, 15, 15))
trainData[name]['patches'].append(patch)
_ = patch_bbox.project(bbox.reproject(landmarkGt[idx]))
trainData[name]['landmarks'].append(_)
for idx, name, padding in types:
logger('writing training data of %s'%name)
patches = np.asarray(trainData[name]['patches'])
landmarks = np.asarray(trainData[name]['landmarks'])
patches = processImage(patches)
shuffle_in_unison_scary(patches, landmarks)
with h5py.File('train/3_%s/%s.h5'%(name, mode), 'w') as h5:
h5['data'] = patches.astype(np.float32)
h5['landmark'] = landmarks.astype(np.float32)
with open('train/3_%s/%s.txt'%(name, mode), 'w') as fd:
fd.write('train/3_%s/%s.h5'%(name, mode))