def inception_v2_ssd(img):
with slim.arg_scope(inception_v2.inception_v2_arg_scope()):
logits, end_point = inception_v2.inception_v2_base(img)
c1 = end_point['Mixed_3c']
c2 = end_point['Mixed_4e']
c3 = end_point['Mixed_5c']
return c1, c2, c3
def inception_v2_ssd(img):
with slim.arg_scope(inception_v2.inception_v2_arg_scope()):
logits, end_point = inception_v2.inception_v2_base(img)
vbs = slim.get_variables_to_restore()
c1 = end_point['Mixed_3c']
c2 = end_point['Mixed_4e']
c3 = end_point['Mixed_5c']
return c1, c2, c3, vbs
def inception_v2_ssd(img,cfg):
with slim.arg_scope(inception_v2.inception_v2_arg_scope()):
logits, end_point = inception_v2.inception_v2_base(img)
Mixed_3c = end_point['Mixed_3c']
Mixed_4e = end_point['Mixed_4e']
cell_11 = end_point['Mixed_5c']
vbs = slim.get_trainable_variables()
cell_11 = tf.image.resize_bilinear(cell_11,size=[32,32])
cell_11 = tf.concat([cell_11,Mixed_4e],axis=3)
cell_7 = tf.image.resize_bilinear(Mixed_4e,size=[64,64])
cell_7 = tf.concat([cell_7,Mixed_3c],axis=3)
cell_11 = slim.conv2d(cell_11,1024,kernel_size=1,activation_fn=slim.nn.relu)
cell_7 = slim.conv2d(cell_7, 512, kernel_size=3, activation_fn=slim.nn.relu)
cell_7 = slim.conv2d(cell_7, 256, kernel_size=1, activation_fn=slim.nn.relu)
cv6 = slim.conv2d(cell_11, 1024, kernel_size=3, rate=6, activation_fn=slim.nn.relu, scope='conv6')
cv7 = slim.conv2d(cv6, 1024, kernel_size=1, activation_fn=slim.nn.relu, scope='conv7')
s = utils.normalize_to_target(cell_7, target_norm_value=12.0, dim=1)
cv8 = slim.conv2d(cv7, 256, kernel_size=1, stride=1, scope='conv8_0')
cv8 = slim.conv2d(cv8, 512, kernel_size=3, stride=2, scope='conv8_1')
cv9 = slim.conv2d(cv8, 128, kernel_size=1, stride=1, scope='conv9_0')
cv9 = slim.conv2d(cv9, 256, kernel_size=3, stride=2, scope='conv9_1')
cv10 = slim.conv2d(cv9, 128, kernel_size=1, stride=1, scope='conv10_0')
cv10 = slim.conv2d(cv10, 256, kernel_size=3, stride=2, scope='conv10_1')
cv11 = slim.conv2d(cv10, 128, kernel_size=1, stride=1, scope='conv11_0')
cv11 = slim.conv2d(cv11, 256, kernel_size=3, stride=2, scope='conv11_1')
source = [s, cv7, cv8, cv9, cv10, cv11]
conf = []
loc = []
for cv, num in zip(source, cfg.Config['aspect_num']):
print(num)
loc.append(slim.conv2d(cv, num * 4, kernel_size=3, stride=1, activation_fn=None))
conf.append(
slim.conv2d(cv, num * cfg.Config['num_classes'], kernel_size=3, stride=1, activation_fn=None))
print(loc)
loc = tf.concat([tf.reshape(o, shape=(cfg.batch_size, -1, 4)) for o in loc], axis=1)
conf = tf.concat([tf.reshape(o, shape=(cfg.batch_size, -1, cfg.Config['num_classes'])) for o in conf],
axis=1)
return loc, conf, vbs
def get_feature_map(self):
input_image = self._processor()
bn_params = {
'is_training': False,
'scale': False,
'decay': 0.9997,
'epsilon': 0.001
}
with slim.arg_scope([slim.conv2d, slim.separable_conv2d],
normalizer_fn=slim.batch_norm,
normalizer_params=bn_params):
_, activations = inception_v2.inception_v2_base(
input_image, final_endpoint='Mixed_4e')
feature_map = activations['Mixed_4e']
return feature_map
def _extract_proposal_features(self, preprocessed_inputs, scope):
"""Extracts first stage RPN features.
Args:
preprocessed_inputs: A [batch, height, width, channels] float32 tensor
representing a batch of images.
scope: A scope name.
Returns:
rpn_feature_map: A tensor with shape [batch, height, width, depth]
activations: A dictionary mapping feature extractor tensor names to
tensors
Raises:
InvalidArgumentError: If the spatial size of `preprocessed_inputs`
(height or width) is less than 33.
ValueError: If the created network is missing the required activation.
"""
preprocessed_inputs.get_shape().assert_has_rank(4)
shape_assert = tf.Assert(
tf.logical_and(
tf.greater_equal(tf.shape(preprocessed_inputs)[1], 33),
tf.greater_equal(tf.shape(preprocessed_inputs)[2], 33)),
['image size must at least be 33 in both height and width.'])
with tf.control_dependencies([shape_assert]):
with tf.variable_scope('InceptionV2',
reuse=self._reuse_weights) as scope:
with _batch_norm_arg_scope(
[slim.conv2d, slim.separable_conv2d],
batch_norm_scale=True,
train_batch_norm=self._train_batch_norm):
_, activations = inception_v2.inception_v2_base(
preprocessed_inputs,
final_endpoint='Mixed_4e',
min_depth=self._min_depth,
depth_multiplier=self._depth_multiplier,
scope=scope)
return tf.concat([
slim.avg_pool2d(activations['Mixed_3c'], [2, 2],
stride=2,
scope='ex_pool3c',
padding='SAME'), activations['Mixed_4e']
],
3,
name='ex_concat'), 0
def inception_v2_ssd(img,cfg):
with slim.arg_scope(inception_v2.inception_v2_arg_scope()):
logits, end_point = inception_v2.inception_v2_base(img)
Mixed_3c = end_point['Mixed_3c']
Mixed_4e = end_point['Mixed_4e']
Mixed_5c = end_point['Mixed_5c']
vbs = slim.get_trainable_variables()
#vbs = None
cell_11 = tf.image.resize_bilinear(Mixed_5c,size=[int(32*(cfg.image_size/512)),int(32*(cfg.image_size/512))])
cell_11 = tf.concat([cell_11,Mixed_4e],axis=3)
cell_7 = tf.image.resize_bilinear(Mixed_4e,size=[int(64*(cfg.image_size/512)),int(64*(cfg.image_size/512))])
cell_7 = tf.concat([cell_7,Mixed_3c],axis=3)
mask_fp = get_mask_fp(Mixed_3c ,Mixed_4e,Mixed_5c)
cell_11 = slim.conv2d(cell_11,1024,kernel_size=1,activation_fn=slim.nn.relu)
cell_7 = slim.conv2d(cell_7, 512, kernel_size=3, activation_fn=slim.nn.relu)
cell_7 = slim.conv2d(cell_7, 256, kernel_size=1, activation_fn=slim.nn.relu)
cv6 = slim.conv2d(cell_11, 1024, kernel_size=3, rate=6, activation_fn=slim.nn.relu, scope='conv6')
cv7 = slim.conv2d(cv6, 1024, kernel_size=1, activation_fn=slim.nn.relu, scope='conv7')
s = utils.normalize_to_target(cell_7, target_norm_value=cfg.norm_value, dim=1)
cv8 = inception(cv7, out_put=512, name='cv8', stride=2)
cv9 = inception(cv8, out_put=256, name='cv9', stride=2)
cv10 = inception(cv9, out_put=256, name='cv10', stride=2)
cv11 = inception(cv10, out_put=256,name= 'cv11', stride=2)
source = [s, cv7, cv8, cv9, cv10, cv11]
conf = []
loc = []
for cv, num in zip(source, cfg.Config['aspect_num']):
loc.append(slim.conv2d(cv, num * 4, kernel_size=3, stride=1, activation_fn=None))
conf.append(
slim.conv2d(cv, num * cfg.Config['num_classes'], kernel_size=3, stride=1, activation_fn=None))
loc = tf.concat([tf.reshape(o, shape=(cfg.batch_size, -1, 4)) for o in loc], axis=1)
conf = tf.concat([tf.reshape(o, shape=(cfg.batch_size, -1, cfg.Config['num_classes'])) for o in conf],
axis=1)
return loc, conf,mask_fp, vbs
def extract_features(self, preprocessed_inputs):
"""Extract features from preprocessed inputs.
Args:
preprocessed_inputs: a [batch, height, width, channels] float tensor
representing a batch of images.
Returns:
feature_maps: a list of tensors where the ith tensor has shape
[batch, height_i, width_i, depth_i]
"""
preprocessed_inputs.get_shape().assert_has_rank(4)
shape_assert = tf.Assert(
tf.logical_and(
tf.greater_equal(tf.shape(preprocessed_inputs)[1], 33),
tf.greater_equal(tf.shape(preprocessed_inputs)[2], 33)),
['image size must at least be 33 in both height and width.'])
feature_map_layout = {
'from_layer': ['Mixed_4c', 'Mixed_5c', '', '', '', ''],
'layer_depth': [-1, -1, 512, 256, 256, 128],
}
with tf.control_dependencies([shape_assert]):
with slim.arg_scope(self._conv_hyperparams):
with tf.variable_scope('InceptionV2',
reuse=self._reuse_weights) as scope:
_, image_features = inception_v2.inception_v2_base(
preprocessed_inputs,
final_endpoint='Mixed_5c',
min_depth=self._min_depth,
depth_multiplier=self._depth_multiplier,
scope=scope)
feature_maps = feature_map_generators.multi_resolution_feature_maps(
feature_map_layout=feature_map_layout,
depth_multiplier=self._depth_multiplier,
min_depth=self._min_depth,
insert_1x1_conv=True,
image_features=image_features)
ret = []
for endpoint_indicator, feature_map in zip(self._endpoints_indicator,
feature_maps.values()):
if endpoint_indicator == '1':
ret.append(feature_map)
return ret
def extract_features(self, inputs):
"""Extracts features from inputs.
This function adds 4 additional feature maps on top of 'Mixed_4c' and
'Mixed_5c' in the base Inception v2 network. For example, if `inputs`
has shape [1, 300, 300, 3], the generated feature maps have the following
shapes:
[
(1, 19, 19, 576), # Mixed_4c
(1, 10, 10, 1024), # Mixed_5c
(1, 5, 5, 512),
(1, 3, 3, 256),
(1, 2, 2, 256),
(1, 1, 1, 128)
]
Args:
inputs: a tensor of shape [batch_size, height, with, channels],
holding the input images.
Returns:
a list of 6 float tensors of shape [batch_size, height, width, channels],
holding feature map tensors to be fed to box predictor.
"""
feature_map_specs_dict = {
'layer_name': ['Mixed_4c', 'Mixed_5c', None, None, None, None],
'layer_depth': [None, None, 512, 256, 256, 128]
}
with slim.arg_scope(self._conv_hyperparams_fn()):
with tf.variable_scope('InceptionV2', reuse=self._reuse_weights) as scope:
_, end_points = inception_v2.inception_v2_base(
inputs,
final_endpoint='Mixed_5c',
min_depth=16,
depth_multiplier=self._depth_multiplier,
scope=scope)
feature_maps = feature_map_generators.ssd_feature_maps(
feature_map_tensor_dict=end_points,
feature_map_specs_dict=feature_map_specs_dict,
depth_multiplier=1, # for depthwise conv in separable_conv2d
use_depthwise=self._use_depthwise,
insert_1x1_conv=True)
feature_map_list = list(feature_maps.values())
return feature_map_list
def extract_features(self, preprocessed_inputs):
"""Extract features from preprocessed inputs.
Args:
preprocessed_inputs: a [batch, height, width, channels] float tensor
representing a batch of images.
Returns:
feature_maps: a list of tensors where the ith tensor has shape
[batch, height_i, width_i, depth_i]
"""
preprocessed_inputs = shape_utils.check_min_image_dim(
33, preprocessed_inputs)
feature_map_layout = {
'from_layer': ['Mixed_4c', 'Mixed_5c', '', '', '',
''][:self._num_layers],
'layer_depth': [-1, -1, 512, 256, 256, 128][:self._num_layers],
'use_explicit_padding':
self._use_explicit_padding,
'use_depthwise':
self._use_depthwise,
}
with slim.arg_scope(self._conv_hyperparams_fn()):
with tf.variable_scope('InceptionV2',
reuse=self._reuse_weights) as scope:
_, image_features = inception_v2.inception_v2_base(
ops.pad_to_multiple(preprocessed_inputs,
self._pad_to_multiple),
final_endpoint='Mixed_5c',
min_depth=self._min_depth,
depth_multiplier=self._depth_multiplier,
scope=scope)
feature_maps = feature_map_generators.multi_resolution_feature_maps(
feature_map_layout=feature_map_layout,
depth_multiplier=self._depth_multiplier,
min_depth=self._min_depth,
insert_1x1_conv=True,
image_features=image_features)
return feature_maps.values()
def extract_features(self, preprocessed_inputs):
"""Extract features from preprocessed inputs.
Args:
preprocessed_inputs: a [batch, height, width, channels] float tensor
representing a batch of images.
Returns:
feature_maps: a list of tensors where the ith tensor has shape
[batch, height_i, width_i, depth_i]
"""
preprocessed_inputs.get_shape().assert_has_rank(4)
shape_assert = tf.Assert(
tf.logical_and(tf.greater_equal(tf.shape(preprocessed_inputs)[1], 33),
tf.greater_equal(tf.shape(preprocessed_inputs)[2], 33)),
['image size must at least be 33 in both height and width.'])
feature_map_layout = {
'from_layer': ['Mixed_4c', 'Mixed_5c', '', '', '', ''],
'layer_depth': [-1, -1, 512, 256, 256, 128],
}
with tf.control_dependencies([shape_assert]):
with slim.arg_scope(self._conv_hyperparams):
with tf.variable_scope('InceptionV2',
reuse=self._reuse_weights) as scope:
_, image_features = inception_v2.inception_v2_base(
ops.pad_to_multiple(preprocessed_inputs, self._pad_to_multiple),
final_endpoint='Mixed_5c',
min_depth=self._min_depth,
depth_multiplier=self._depth_multiplier,
scope=scope)
feature_maps = feature_map_generators.multi_resolution_feature_maps(
feature_map_layout=feature_map_layout,
depth_multiplier=self._depth_multiplier,
min_depth=self._min_depth,
insert_1x1_conv=True,
image_features=image_features)
return feature_maps.values()
def _extract_proposal_features(self, preprocessed_inputs, scope):
"""Extracts first stage RPN features.
Args:
preprocessed_inputs: A [batch, height, width, channels] float32 tensor
representing a batch of images.
scope: A scope name.
Returns:
rpn_feature_map: A tensor with shape [batch, height, width, depth]
activations: A dictionary mapping feature extractor tensor names to
tensors
Raises:
InvalidArgumentError: If the spatial size of `preprocessed_inputs`
(height or width) is less than 33.
ValueError: If the created network is missing the required activation.
"""
preprocessed_inputs.get_shape().assert_has_rank(4)
shape_assert = tf.Assert(
tf.logical_and(tf.greater_equal(tf.shape(preprocessed_inputs)[1], 33),
tf.greater_equal(tf.shape(preprocessed_inputs)[2], 33)),
['image size must at least be 33 in both height and width.'])
with tf.control_dependencies([shape_assert]):
with tf.variable_scope('InceptionV2',
reuse=self._reuse_weights) as scope:
with _batch_norm_arg_scope([slim.conv2d, slim.separable_conv2d],
batch_norm_scale=True,
train_batch_norm=self._train_batch_norm):
_, activations = inception_v2.inception_v2_base(
preprocessed_inputs,
final_endpoint='Mixed_4e',
min_depth=self._min_depth,
depth_multiplier=self._depth_multiplier,
scope=scope)
return activations['Mixed_4e'], activations
def model(img):
with slim.arg_scope(inception_v2.inception_v2_arg_scope()):
with slim.arg_scope([slim.batch_norm], is_training=True):
with slim.arg_scope([slim.conv2d], trainable=True):
logits, end_point = inception_v2.inception_v2_base(img)
c1 = end_point['Mixed_3c']
c2 = end_point['Mixed_4e']
c3 = end_point['Mixed_5c']
vbs = slim.get_variables_to_restore()
c3 = slim.conv2d(c3, 256, 1, 1, activation_fn=None)
c2 = slim.conv2d(c2, 256, 1, 1,
activation_fn=None) + tf.image.resize_bilinear(
c3, size=tf.shape(c2)[1:3])
c1 = slim.conv2d(c1, 256, 1, 1,
activation_fn=None) + tf.image.resize_bilinear(
c2, size=tf.shape(c1)[1:3])
return c1, c2, c3, vbs
def extract_features(self, preprocessed_inputs):
"""Extract features from preprocessed inputs.
Args:
preprocessed_inputs: a [batch, height, width, channels] float tensor
representing a batch of images.
Returns:
feature_maps: a list of tensors where the ith tensor has shape
[batch, height_i, width_i, depth_i]
"""
preprocessed_inputs = shape_utils.check_min_image_dim(
33, preprocessed_inputs)
feature_map_layout = {
'from_layer': ['Mixed_4c', 'Mixed_5c', '', '', '', ''],
'layer_depth': [-1, -1, 512, 256, 256, 128],
'use_explicit_padding': self._use_explicit_padding,
'use_depthwise': self._use_depthwise,
}
with slim.arg_scope(self._conv_hyperparams_fn()):
with tf.variable_scope('InceptionV2',
reuse=self._reuse_weights) as scope:
_, image_features = inception_v2.inception_v2_base(
ops.pad_to_multiple(preprocessed_inputs, self._pad_to_multiple),
final_endpoint='Mixed_5c',
min_depth=self._min_depth,
depth_multiplier=self._depth_multiplier,
scope=scope)
feature_maps = feature_map_generators.multi_resolution_feature_maps(
feature_map_layout=feature_map_layout,
depth_multiplier=self._depth_multiplier,
min_depth=self._min_depth,
insert_1x1_conv=True,
image_features=image_features)
return feature_maps.values()
def extract_features(self, preprocessed_inputs):
"""Extract features from preprocessed inputs.
Args:
preprocessed_inputs: a [batch, height, width, channels] float tensor
representing a batch of images.
Returns:
feature_maps: a list of tensors where the ith tensor has shape
[batch, height_i, width_i, depth_i]
"""
# Make sure that input is in correct format with rank 4.
preprocessed_inputs.get_shape().assert_has_rank(4)
shape_assert = tf.Assert(
tf.logical_and(
tf.greater_equal(tf.shape(preprocessed_inputs)[1], 33),
tf.greater_equal(tf.shape(preprocessed_inputs)[2], 33)),
['image size must at least be 33 in both height and width.'])
with tf.control_dependencies([shape_assert]):
with slim.arg_scope(self._conv_hyperparams):
with tf.variable_scope('InceptionV2',
reuse=self._reuse_weights) as scope:
_, image_features = inception_v2.inception_v2_base(
ops.pad_to_multiple(preprocessed_inputs,
self._pad_to_multiple),
final_endpoint='Mixed_5c',
min_depth=self._min_depth,
depth_multiplier=self._depth_multiplier,
use_separable_conv=False,
scope=scope)
# 2. STDN version + combine mode
image_features = combine_and_scale_transfer_module_v1(
image_features, combine_mode=0)
# return a list of feature maps
return image_features.values()
def _extract_first_stage_features(self, inputs):
"""Extracts first stage features for RPN proposal prediction and
for ROI pooling.
Args:
inputs: float tensor of shape [batch_size, height, width, depth].
Returns:
shared_feature_map: float tensor of shape
[batch_size, height_out, width_out, depth_out].
"""
with tf.variable_scope('InceptionV2', reuse=self._reuse_weights) as scope:
with slim.arg_scope([slim.conv2d, slim.separable_conv2d],
# is_training
normalizer_fn=slim.batch_norm,
normalizer_params=self._batch_norm_params):
_, end_points = inception_v2.inception_v2_base(
inputs,
final_endpoint='Mixed_4e',
min_depth=16,
depth_multiplier=self._depth_multiplier,
scope=scope)
return end_points['Mixed_4e']
def _extract_proposal_features(self, preprocessed_inputs, scope):
"""Extracts first stage RPN features.
Args:
preprocessed_inputs: A [batch, height, width, channels] float32 tensor
representing a batch of images.
scope: A scope name.
Returns:
rpn_feature_map: A tensor with shape [batch, height, width, depth]
activations: A dictionary mapping feature extractor tensor names to
tensors
Raises:
InvalidArgumentError: If the spatial size of `preprocessed_inputs`
(height or width) is less than 33.
ValueError: If the created network is missing the required activation.
"""
preprocessed_inputs.get_shape().assert_has_rank(4)
shape_assert = tf.Assert(
tf.logical_and(
tf.greater_equal(tf.shape(preprocessed_inputs)[1], 33),
tf.greater_equal(tf.shape(preprocessed_inputs)[2], 33)),
['image size must at least be 33 in both height and width.'])
depth = lambda d: max(int(d * self._depth_multiplier), self._min_depth)
trunc_normal = lambda stddev: tf.truncated_normal_initializer(
0.0, stddev)
#add convolution autoencoder
encoder_1_conv = slim.conv2d(preprocessed_inputs,
depth(64), [3, 3],
weights_initializer=trunc_normal(0.09),
scope='encoder_c1_conv')
#encoder_1_pool = slim.max_pool2d(encoder_1_conv, [2, 2], stride=2,
# scope='encoder_c1_pool')
#encoder_1_dropout1 = slim.dropout(encoder_1_pool, 0.7, scope='encoder_c1_dropout1')
encoder_2_conv = slim.conv2d(encoder_1_conv,
depth(128), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='encoder_c2_conv')
encoder_3_conv = slim.conv2d(encoder_2_conv,
depth(128), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='encoder_c3_conv')
#decoder
decoder_3_deconv = slim.conv2d_transpose(
encoder_3_conv,
depth(128), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='decoder_c3_deconv')
decoder_2_deconv = slim.conv2d_transpose(
decoder_3_deconv,
depth(128), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='decoder_c2_deconv')
decoder_1_deconv = slim.conv2d_transpose(
decoder_2_deconv,
depth(1), [3, 3],
weights_initializer=trunc_normal(0.09),
scope='decoder_c1_deconv')
#visualize decoder output
#vis_decoder = tf.slice(decoder_1_deconv,(0,0,0,0),(1,-1,-1,-1))
#print("vis_decoder:",vis_decoder)
#vis_decoder = tf.reshape(vis_decoder,(256,256,1))
#print("vis_decoder2:",vis_decoder)
#vis_decoder = tf.transpose(vis_decoder,(2,0,3,1))
#vis_decoder = tf.reshape(vis_decoder,(1,256,256,1))
#image_matrix = vis_decoder.eval()
#print("image_matrix:",image_matrix)
#tf.summary.image('decoder_visualized',tf.expand_dims(vis_decoder, 0)
with tf.control_dependencies([shape_assert]):
with tf.variable_scope('InceptionV2',
reuse=self._reuse_weights) as scope:
with _batch_norm_arg_scope(
[slim.conv2d, slim.separable_conv2d],
batch_norm_scale=True,
train_batch_norm=self._train_batch_norm):
_, activations = inception_v2.inception_v2_base(
decoder_1_deconv,
final_endpoint='Mixed_4e',
min_depth=self._min_depth,
depth_multiplier=self._depth_multiplier,
scope=scope)
return activations['Mixed_4e'], activations
def _extract_proposal_features(self, preprocessed_inputs, scope):
"""Extracts first stage RPN features.
Args:
preprocessed_inputs: A [batch, height, width, channels] float32 tensor
representing a batch of images.
scope: A scope name.
Returns:
rpn_feature_map: A tensor with shape [batch, height, width, depth]
activations: A dictionary mapping feature extractor tensor names to
tensors
Raises:
InvalidArgumentError: If the spatial size of `preprocessed_inputs`
(height or width) is less than 33.
ValueError: If the created network is missing the required activation.
"""
preprocessed_inputs.get_shape().assert_has_rank(4)
shape_assert = tf.Assert(
tf.logical_and(
tf.greater_equal(tf.shape(preprocessed_inputs)[1], 33),
tf.greater_equal(tf.shape(preprocessed_inputs)[2], 33)),
['image size must at least be 33 in both height and width.'])
depth = lambda d: max(int(d * self._depth_multiplier), self._min_depth)
trunc_normal = lambda stddev: tf.truncated_normal_initializer(
0.0, stddev)
'''
Add a multi branch encoder with different dilation
'''
'''
Block 1
'''
encoder_1_conv = slim.conv2d(preprocessed_inputs,
depth(32), [3, 3],
weights_initializer=trunc_normal(0.09),
scope='encoder_c1_conv')
encoder_2_conv = slim.conv2d(encoder_1_conv,
depth(64), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='encoder_c2_conv')
encoder_3_conv = slim.conv2d(encoder_2_conv,
depth(128), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='encoder_c3_conv')
'''
Block 2
'''
encoder_1_conv_dilated_1 = slim.conv2d(
preprocessed_inputs,
depth(32), [3, 3],
rate=3,
weights_initializer=trunc_normal(0.09),
scope='encoder_1_conv_dilated_1')
encoder_2_conv_dilated_1 = slim.conv2d(
encoder_1_conv_dilated_1,
depth(64), [5, 5],
rate=3,
weights_initializer=trunc_normal(0.09),
scope='encoder_2_conv_dilated_1')
encoder_3_conv_dilated_1 = slim.conv2d(
encoder_2_conv_dilated_1,
depth(128), [5, 5],
rate=3,
weights_initializer=trunc_normal(0.09),
scope='encoder_3_conv_dilated_1')
'''
Block 3
'''
encoder_1_conv_dilated_2 = slim.conv2d(
preprocessed_inputs,
depth(32), [3, 3],
rate=5,
weights_initializer=trunc_normal(0.09),
scope='encoder_1_conv_dilated_2')
encoder_2_conv_dilated_2 = slim.conv2d(
encoder_1_conv_dilated_2,
depth(64), [5, 5],
rate=5,
weights_initializer=trunc_normal(0.09),
scope='encoder_2_conv_dilated_2')
encoder_3_conv_dilated_2 = slim.conv2d(
encoder_2_conv_dilated_2,
depth(128), [5, 5],
rate=5,
weights_initializer=trunc_normal(0.09),
scope='encoder_3_conv_dilated_2')
encoder_concat = tf.concat([
encoder_3_conv, encoder_3_conv_dilated_1, encoder_3_conv_dilated_2
],
axis=3)
#decoder
decoder_3_deconv = slim.conv2d_transpose(
encoder_concat,
depth(64), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='decoder_c3_deconv')
decoder_2_deconv = slim.conv2d_transpose(
decoder_3_deconv,
depth(32), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='decoder_c2_deconv')
decoder_1_deconv = slim.conv2d_transpose(
decoder_2_deconv,
depth(1), [3, 3],
weights_initializer=trunc_normal(0.09),
scope='decoder_c1_deconv')
with tf.control_dependencies([shape_assert]):
with tf.variable_scope('InceptionV2',
reuse=self._reuse_weights) as scope:
with _batch_norm_arg_scope(
[slim.conv2d, slim.separable_conv2d],
batch_norm_scale=True,
train_batch_norm=self._train_batch_norm):
_, activations = inception_v2.inception_v2_base(
decoder_1_deconv,
final_endpoint='Mixed_4e',
min_depth=self._min_depth,
depth_multiplier=self._depth_multiplier,
scope=scope)
return activations['Mixed_4e'], activations
def network_fn(inputs):
# return transformer_factory.transform(inputs, BATCH_PER_GPU, NUM_STN, (224, 224), NUM_CLASSES, FLAGS.weight_decay, True)
end_points = {}
# with slim.arg_scope([slim.batch_norm, slim.dropout], is_training=True):
# with slim.arg_scope(inception_v3_arg_scope(weight_decay=FLAGS.weight_decay)):
with slim.arg_scope([slim.batch_norm, slim.dropout],
is_training=is_training):
with slim.arg_scope(inception_v3_arg_scope(weight_decay=weight_decay)):
with tf.variable_scope("loc") as scope:
with tf.variable_scope("net") as scope2:
# _, _end_points = inception_resnet_v2.inception_resnet_v2(inputs, num_classes=2, is_training=True, scope = scope2)
loc_net, _ = inception_v2.inception_v2_base(inputs,
scope=scope2)
# loc_net = _end_points['Conv2d_7b_1x1']
loc_net = slim.conv2d(loc_net, 128, [1, 1], scope='Loc_1x1')
default_kernel_size = [14, 14]
# kernel_size = _reduced_kernel_size_for_small_input(loc_net, default_kernel_size)
loc_net = slim.conv2d(loc_net,
128,
loc_net.get_shape()[1:3],
padding='VALID',
activation_fn=tf.nn.tanh,
scope='Loc_fc1')
loc_net = slim.flatten(loc_net)
iv = 4.
initial = np.array([iv, 0, iv, 0] * NUM_STN, dtype=np.float32)
b_fc_loc = tf.get_variable(
"Loc_fc_b",
shape=[4 * NUM_STN],
initializer=init_ops.constant_initializer(initial),
dtype=dtypes.float32)
W_fc_loc = tf.get_variable(
"Loc_fc_W",
shape=[128, 4 * NUM_STN],
initializer=init_ops.constant_initializer(
np.zeros((128, 4 * NUM_STN))),
dtype=dtypes.float32)
theta = tf.nn.tanh(tf.matmul(loc_net, W_fc_loc) + b_fc_loc)
_finals = []
for i in xrange(NUM_STN):
scope_name = "stn%d" % i
with tf.variable_scope(scope_name) as scope1:
_theta = tf.slice(theta, [0, 4 * i], [-1, 4 * (i + 1)])
# loc_net = slim.conv2d(loc_net, 6, [1,1], activation_fn=tf.nn.tanh, scope='Loc_fc', biases_initializer = init_ops.constant_initializer([4.0,0.0,0.0,0.0,4.0,0.0]*128,dtype=dtypes.float32))
# loc_net = slim.conv2d(loc_net, 6, [1,1], activation_fn=tf.nn.tanh, scope='Loc_fc', biases_initializer = init_ops.constant_initializer([4.0],dtype=dtypes.float32))
# loc_net = slim.flatten(loc_net)
stn_output_size = (STN_OUT_SIZE, STN_OUT_SIZE)
x = transformer(inputs, _theta, stn_output_size)
x.set_shape([
BATCH_PER_GPU, stn_output_size[0], stn_output_size[1],
3
])
# x.set_shape(tf.shape(inputs))
# tf.reshape(x, tf.shape(inputs))
end_points['x'] = x
# with tf.variable_scope("net") as scope2:
# return inception_resnet_v2.inception_resnet_v2(x, num_classes=NUM_CLASSES, is_training=True, scope = scope2)
with tf.variable_scope("net") as scope2:
net, _ = inception_v2.inception_v2_base(x,
scope=scope2)
kernel_size = _reduced_kernel_size_for_small_input(
net, [7, 7])
net = slim.avg_pool2d(net,
kernel_size,
padding='VALID',
scope='AvgPool_1a')
net = slim.dropout(net, keep_prob=0.7, scope='Dropout_1b')
_finals.append(net)
with tf.variable_scope('Logits'):
net = tf.concat(axis=3, values=_finals)
logits = slim.conv2d(net,
NUM_CLASSES, [1, 1],
activation_fn=None,
normalizer_fn=None,
scope='Conv2d_1c_1x1')
logits = tf.squeeze(logits, [1, 2], name='SpatialSqueeze')
predictions = slim.softmax(logits, scope='Predictions')
end_points['Predictions'] = predictions
logits_a = slim.conv2d(net,
NUM_ATTRIBS, [1, 1],
activation_fn=None,
normalizer_fn=None,
scope='Conv2d_1c_1x1_a')
logits_a = tf.squeeze(logits_a, [1, 2],
name='SpatialSqueeze_a')
predictions_a = slim.sigmoid(logits_a, scope='Predictions_a')
end_points['Predictions_a'] = predictions_a
return logits, logits_a, end_points
def extract_features(self, preprocessed_inputs):
"""Extract features from preprocessed inputs.
Args:
preprocessed_inputs: a [batch, height, width, channels] float tensor
representing a batch of images.
Returns:
feature_maps: a list of tensors where the ith tensor has shape
[batch, height_i, width_i, depth_i]
"""
preprocessed_inputs = shape_utils.check_min_image_dim(
33, preprocessed_inputs)
depth = lambda d: max(int(d * self._depth_multiplier), self._min_depth)
trunc_normal = lambda stddev: tf.truncated_normal_initializer(
0.0, stddev)
#add convolution autoencoder
encoder_1_conv = slim.conv2d(preprocessed_inputs,
depth(64), [3, 3],
weights_initializer=trunc_normal(0.09),
scope='encoder_c1_conv')
#encoder_1_pool = slim.max_pool2d(encoder_1_conv, [2, 2], stride=2,
# scope='encoder_c1_pool')
#encoder_1_dropout1 = slim.dropout(encoder_1_pool, 0.7, scope='encoder_c1_dropout1')
encoder_2_conv = slim.conv2d(encoder_1_conv,
depth(128), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='encoder_c2_conv')
encoder_3_conv = slim.conv2d(encoder_2_conv,
depth(128), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='encoder_c3_conv')
#decoder
decoder_3_deconv = slim.conv2d_transpose(
encoder_3_conv,
depth(128), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='decoder_c3_deconv')
decoder_2_deconv = slim.conv2d_transpose(
decoder_3_deconv,
depth(128), [5, 5],
weights_initializer=trunc_normal(0.09),
scope='decoder_c2_deconv')
decoder_1_deconv = slim.conv2d_transpose(
decoder_2_deconv,
depth(1), [3, 3],
weights_initializer=trunc_normal(0.09),
scope='decoder_c1_deconv')
feature_map_layout = {
'from_layer': ['Mixed_4c', 'Mixed_5c', '', '', '', ''],
'layer_depth': [-1, -1, 512, 256, 256, 128],
'use_explicit_padding': self._use_explicit_padding,
'use_depthwise': self._use_depthwise,
}
with slim.arg_scope(self._conv_hyperparams_fn()):
with tf.variable_scope('InceptionV2',
reuse=self._reuse_weights) as scope:
_, image_features = inception_v2.inception_v2_base(
ops.pad_to_multiple(decoder_1_deconv,
self._pad_to_multiple),
final_endpoint='Mixed_5c',
min_depth=self._min_depth,
depth_multiplier=self._depth_multiplier,
scope=scope)
feature_maps = feature_map_generators.multi_resolution_feature_maps(
feature_map_layout=feature_map_layout,
depth_multiplier=self._depth_multiplier,
min_depth=self._min_depth,
insert_1x1_conv=True,
image_features=image_features)
return feature_maps.values()