def resnet50_block(inputs, num_pyramid):
bn_axis = 3 # TODO
outputs = []
x = layers.ZeroPadding2D(padding=(3, 3), name='conv1_pad')(inputs)
x = layers.Conv2D(64, (7, 7),
strides=(2, 2),
padding='valid',
kernel_initializer='he_normal',
name='conv1')(x)
x = batch_norm(axis=bn_axis, name='bn_conv1')(x)
x = layers.Activation('relu')(x)
x = layers.ZeroPadding2D(padding=(1, 1), name='pool1_pad')(x)
x = layers.MaxPooling2D((3, 3), strides=(2, 2))(x)
x = conv_block(x, 3, [64, 64, 256], stage=2, block='a', strides=(1, 1))
x = identity_block(x, 3, [64, 64, 256], stage=2, block='b')
x = identity_block(x, 3, [64, 64, 256], stage=2, block='c')
if num_pyramid > 3:
outputs.append(x) # 1/4
x = conv_block(x, 3, [128, 128, 512], stage=3, block='a')
x = identity_block(x, 3, [128, 128, 512], stage=3, block='b')
x = identity_block(x, 3, [128, 128, 512], stage=3, block='c')
x = identity_block(x, 3, [128, 128, 512], stage=3, block='d')
if num_pyramid > 2:
outputs.append(x) # 1/8
x = conv_block(x, 3, [256, 256, 1024], stage=4, block='a')
x = identity_block(x, 3, [256, 256, 1024], stage=4, block='b')
x = identity_block(x, 3, [256, 256, 1024], stage=4, block='c')
x = identity_block(x, 3, [256, 256, 1024], stage=4, block='d')
x = identity_block(x, 3, [256, 256, 1024], stage=4, block='e')
x = identity_block(x, 3, [256, 256, 1024], stage=4, block='f')
if num_pyramid > 1:
outputs.append(x) # 1/16
x = conv_block(x, 3, [512, 512, 2048], stage=5, block='a')
x = identity_block(x, 3, [512, 512, 2048], stage=5, block='b')
x = identity_block(x, 3, [512, 512, 2048], stage=5, block='c')
outputs.append(x) # 1/32
return outputs
def Discriminator(img_size=img_size,channels=channels):
initializer=tf.random_normal_initializer(0.,.02)
inp=tkl.Input(shape=[img_size,img_size,channels],
name='input_img')
tar=tkl.Input(shape=[img_size,img_size,channels],
name='target_img')
x=tkl.concatenate([inp,tar])
ds1=downsample(64,4,False)(x)
ds2=downsample(192,4)(ds1)
ds3=downsample(256,4)(ds2)
zero_pad1=tkl.ZeroPadding2D()(ds3)
conv=tkl.Conv2D(
512,4,strides=1,kernel_initializer=initializer,
use_bias=False)(zero_pad1)
batchnorm1=tkl.BatchNormalization()(conv)
leaky_relu=tkl.LeakyReLU()(batchnorm1)
zero_pad2=tkl.ZeroPadding2D()(leaky_relu)
last=tkl.Conv2D(
1,4,strides=1,kernel_initializer=initializer)(zero_pad2)
return tf.keras.Model(inputs=[inp,tar],outputs=last)
