def Pelee(net, from_layer='data', use_batchnorm=False):
PeleeNetBody(net, from_layer)
add_extra_layers_pelee(net, use_batchnorm=use_batchnorm, prefix='ext1/fe')
fpn_block(net)
raw_source_layers = ['p1', 'p2', 'p3', 'p4', 'ext1/fe2_2', 'ext1/fe3_2']
# add_res_prediction_layers
last_base_layer = 'fpn_tb'
for i, from_layer in enumerate(raw_source_layers):
out_layer = '{}/ext/pm{}'.format(last_base_layer, i + 1)
res_block(net, from_layer, 256, out_layer, stride=1, use_bn=True)
return net
def VGG_RUN(net, from_layer='data', use_batchnorm=False):
VGGNetBody(net, from_layer=from_layer, fully_conv=True, reduced=True, dilated=True, dropout=False)
add_extra_layers_default(net, use_batchnorm=False)
# add_res_prediction_layers
raw_source_layers=VGG_SSD.mbox_source_layers
for i in range(len(raw_source_layers)):
name = 'ext/pm{}'.format(i+1)
res_block(net, raw_source_layers[i], 256, name, stride=1, use_bn=True)
return net
def Pelee(net, from_layer='data', use_batchnorm=False):
PeleeNetBody(net, from_layer)
add_extra_layers_pelee(net, use_batchnorm=use_batchnorm, prefix='ext1/fe')
raw_source_layers = ['stage3_tb', 'stage4_tb','ext1/fe1_2', 'ext1/fe2_2','ext1/fe3_2']
# add_res_prediction_layers
last_base_layer = 'stage4_tb'
for i, from_layer in enumerate(raw_source_layers):
out_layer = '{}/ext/pm{}'.format(last_base_layer, i+2)
res_block(net, from_layer, 256, out_layer, stride=1, use_bn=True)
return net
def generator_model():
"""Build generator architecture."""
# Current version : ResNet block
inputs = Input(shape=image_shape)
x = ReflectionPadding2D((3, 3))(inputs)
x = Conv2D(filters=ngf, kernel_size=(7, 7), padding='valid')(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
# Increase filter number
n_downsampling = 2
for i in range(n_downsampling):
mult = 2**i
x = Conv2D(filters=ngf * mult * 2,
kernel_size=(3, 3),
strides=2,
padding='same')(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
# Apply 9 ResNet blocks
mult = 2**n_downsampling
for i in range(n_blocks_gen):
x = res_block(x, ngf * mult, use_dropout=True)
# Decrease filter number to 3 (RGB)
for i in range(n_downsampling):
mult = 2**(n_downsampling - i)
x = Conv2DTranspose(filters=int(ngf * mult / 2),
kernel_size=(3, 3),
strides=2,
padding='same')(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x = ReflectionPadding2D((3, 3))(x)
x = Conv2D(filters=output_nc, kernel_size=(7, 7), padding='valid')(x)
x = Activation('tanh')(x)
# Add direct connection from input to output and recenter to [-1, 1]
outputs = Add()([x, inputs])
outputs = Lambda(lambda z: z / 2)(outputs)
model = Model(inputs=inputs, outputs=outputs, name='Generator')
return model
def generator_model():
"""Build generator architecture."""
# Current version : ResNet block
inputs = Input(shape=image_shape)
x = ReflectionPadding2D((3, 3))(inputs)
x = Conv2D(filters=ngf, kernel_size=(7, 7), padding='valid')(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
n_downsampling = 3
for i in range(n_downsampling):
mult = 2**i
x = Conv2D(filters=ngf * mult * 2,
kernel_size=(3, 3),
strides=2,
padding='same')(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
mult = 2**n_downsampling
for i in range(n_blocks_gen):
x = res_block(x, ngf * mult, use_dropout=True)
for i in range(n_downsampling):
mult = 2**(n_downsampling - i)
x = Conv2DTranspose(filters=int(ngf * mult / 2),
kernel_size=(3, 3),
strides=2,
padding='same')(x)
# x = BatchNormalization()(x)
x = Activation('relu')(x)
x = ReflectionPadding2D((3, 3))(x)
x = Conv2D(filters=output_nc, kernel_size=(7, 7), padding='valid')(x)
x = Activation('tanh')(x)
outputs = x
# outputs = Add()([x, inputs])
# outputs = Lambda(lambda z: K.clip(z, -1, 1))(x)
# outputs = Lambda(lambda z: z/2)(outputs)
model = Model(inputs=inputs, outputs=outputs, name='Generator')
plot_model(model, to_file='generator.png', show_shapes=True)
return model
def generator_model():
"""生成器模型
"""
inputs = Input(Config.input_shape_generator)
x = ReflectionPadding2D((3, 3))(inputs)
print(x.shape)
x = Conv2D(filters=Config.ngf, kernel_size=(7, 7), padding="valid")(x)
x = BatchNormalization()(x)
x = Activation("relu")(x)
n_downsampling = 2
for i in range(n_downsampling):
mulit = 2**i
x = Conv2D(filters=Config.ngf * mulit * 2,
kernel_size=(3, 3),
strides=2,
padding="same")(x)
x = BatchNormalization()(x)
x = Activation("relu")(x)
mulit = 2**n_downsampling
for i in range(Config.n_blocks_gen):
x = res_block(x, Config.ngf * mulit, use_dropout=True)
for i in range(n_downsampling):
mulit = 2**(n_downsampling - i)
x = Conv2DTranspose(filters=int(Config.ngf * mulit / 2),
kernel_size=(3, 3),
strides=2,
padding="same")(x)
x = BatchNormalization()(x)
x = Activation("relu")(x)
x = ReflectionPadding2D(padding=(3, 3))(x)
x = Conv2D(filters=Config.output_nc,
kernel_size=(7, 7),
padding="valid")(x)
x = Activation("tanh")(x)
# 输出
outputs = Add()([inputs, x])
outputs = Lambda(lambda z: z / 2)(outputs)
print("generator : ", outputs.shape)
model = Model(inputs=inputs, outputs=outputs, name="Generator")
return model
def generator_model(args, inputs, istrain, reuse):
"""Build generator architecture."""
# inputs: tensor with shape [bn, 256,256, 1]
# inputs = Input(shape=input_shape_generator)
with tf.variable_scope('gen_', reuse=reuse):
x = ReflectionPadding2D((3, 3))(inputs)
x = Conv2D(filters=ngf, kernel_size=(7, 7), padding='valid')(x)
x = batch_norm(x, "bn1", is_train=istrain)
x = Activation('relu')(x)
# x = MaxPooling2D((2, 2), padding='same')(x)e')(x)
# x = Conv2D(filters=ngf, kernel_size=(7,7), padding='same')(x)
# x = batch_norm(x, "bn2", is_train=istrain)
# x = Activation('relu')(x)
n_downsampling = 2
for i in range(n_downsampling):
mult = 2**i
# x = ReflectionPadding2D((2, 2))(x)
if args.max_pooling == False:
x = Conv2D(filters=ngf * mult * 2,
kernel_size=(3, 3),
strides=2,
padding='valid')(x)
else:
x = Conv2D(filters=ngf * mult * 2,
kernel_size=(3, 3),
strides=1,
padding='valid')(x)
x = MaxPooling2D((2, 2), padding='valid')(x)
# x = BatchNormalization()(x, training=istrain)
x = batch_norm(x, "down_bn_" + str(i), is_train=istrain)
tf.summary.histogram('before_active', x)
x = Activation('relu')(x)
tf.summary.histogram('after_activate', x)
mult = 2**n_downsampling
for i in range(n_blocks_gen):
x = res_block(x, ngf * mult, use_dropout=True)
# for i in range(n_downsampling):
# mult = 2**(n_downsampling - i)
# x = UpSampling2D()(x)
# x = Conv2D(filters=int(ngf * mult / 2),kernel_size=(3,3),padding='same')(x)
## x = Conv2DTranspose(filters=int(ngf * mult / 2), kernel_size=(3, 3), strides=2, padding='same')(x)
## x = BatchNormalization()(x, training=istrain)
# x = batch_norm(x, "up_bn_"+str(i), is_train=istrain)
# x = LeakyReLU(alpha=0.3)(x)
x = Conv2D(filters=2, kernel_size=(5, 5), padding='same')(x)
x = batch_norm(x, "final", is_train=istrain)
wrap = Activation('sigmoid')(x)
wrap = tf.multiply(tf.add(wrap, -0.5), 8)
# dense layer
dense = tf.layers.flatten(wrap)
output_size = 128
output_size = args.final_layer # we use the args value here to decide the final layer number
# output_size1 = 16
dense_out = tf.layers.dense(inputs=dense, units=output_size * 2)
# dense_out1 = tf.layers.dense(inputs=dense_out, units=output_size1*2)
x_mean = tf.reshape(dense_out, [-1, output_size, 2])
# x_mean = Conv2D(filters=2, kernel_size=(1,256), padding='valid')(wrap)
# x_layer = wrap[...,0]
# x_mean = tf.reduce_max(wrap, axis=2)
x_mean = tf.expand_dims(x_mean, 2)
wrap = tf.tile(x_mean, multiples=[1, 1, output_size, 1])
wrap = bicubic_interp_2d(wrap, imsize)
outputs = Lambda(WarpST_one,
arguments={
'inputs': inputs,
'name': str(random.random())
})(wrap)
return outputs, wrap[:, :, 0, :]
