Pix2Pix
Image-to-Image Translation with Conditional Adversarial Networks
We investigate conditional adversarial networks as a general-purpose solution to image-to-image translation problems. These networks not only learn the mapping from input image to output image, but also learn a loss function to train this mapping. This makes it possible to apply the same generic approach to problems that traditionally would require very different loss formulations. We demonstrate that this approach is effective at synthesizing photos from label maps, reconstructing objects from edge maps, and colorizing images, among other tasks. Indeed, since the release of the pix2pix software associated with this paper, a large number of internet users (many of them artists) have posted their own experiments with our system, further demonstrating its wide applicability and ease of adoption without the need for parameter tweaking. As a community, we no longer hand-engineer our mapping functions, and this work suggests we can achieve reasonable results without hand-engineering our loss functions either.
Paper: Link
Implementations
Pix2Pix implementation with Tensorflow 2.0 and Keras
Build the Generator
- The architecture of generator is a modified U-Net
- Each block in the encoder is (Conv -> Batchnorm -> Leaky ReLU)
- Each block in the decoder is (Transposed Conv -> Batchnorm -> Dropout(applied to the first 3 blocks) -> ReLU)
- There are skip connections between the encoder and decoder (as in U-Net)
def downsample(filters, size, apply_batchnorm=True):
initializer = tf.random_normal_initializer(0., 0.02)
result = tf.keras.Sequential()
result.add(
tf.keras.layers.Conv2D(filters, size, strides=2, padding='same',
kernel_initializer=initializer, use_bias=False))
if apply_batchnorm:
result.add(tf.keras.layers.BatchNormalization())
result.add(tf.keras.layers.LeakyReLU())
return resultdown_model = downsample(3, 4)
down_result = down_model(tf.expand_dims(inp, 0))
print (down_result.shape)(1, 128, 128, 3)def upsample(filters, size, apply_dropout=False):
initializer = tf.random_normal_initializer(0., 0.02)
result = tf.keras.Sequential()
result.add(
tf.keras.layers.Conv2DTranspose(filters, size, strides=2,
padding='same',
kernel_initializer=initializer,
use_bias=False))
result.add(tf.keras.layers.BatchNormalization())
if apply_dropout:
result.add(tf.keras.layers.Dropout(0.5))
result.add(tf.keras.layers.ReLU())
return resultup_model = upsample(3, 4)
up_result = up_model(down_result)
print (up_result.shape)(1, 256, 256, 3)def Generator():
down_stack = [
downsample(64, 4, apply_batchnorm=False), # (bs, 128, 128, 64)
downsample(128, 4), # (bs, 64, 64, 128)
downsample(256, 4), # (bs, 32, 32, 256)
downsample(512, 4), # (bs, 16, 16, 512)
downsample(512, 4), # (bs, 8, 8, 512)
downsample(512, 4), # (bs, 4, 4, 512)
downsample(512, 4), # (bs, 2, 2, 512)
downsample(512, 4), # (bs, 1, 1, 512)
]
up_stack = [
upsample(512, 4, apply_dropout=True), # (bs, 2, 2, 1024)
upsample(512, 4, apply_dropout=True), # (bs, 4, 4, 1024)
upsample(512, 4, apply_dropout=True), # (bs, 8, 8, 1024)
upsample(512, 4), # (bs, 16, 16, 1024)
upsample(256, 4), # (bs, 32, 32, 512)
upsample(128, 4), # (bs, 64, 64, 256)
upsample(64, 4), # (bs, 128, 128, 128)
]
initializer = tf.random_normal_initializer(0., 0.02)
last = tf.keras.layers.Conv2DTranspose(OUTPUT_CHANNELS, 4,
strides=2,
padding='same',
kernel_initializer=initializer,
activation='tanh') # (bs, 256, 256, 3)
concat = tf.keras.layers.Concatenate()
inputs = tf.keras.layers.Input(shape=[None,None,3])
x = inputs
# Downsampling through the model
skips = []
for down in down_stack:
x = down(x)
skips.append(x)
skips = reversed(skips[:-1])
# Upsampling and establishing the skip connections
for up, skip in zip(up_stack, skips):
x = up(x)
x = concat([x, skip])
x = last(x)
return tf.keras.Model(inputs=inputs, outputs=x)Build the Discriminator
- The Discriminator is a PatchGAN
- Each block in the discriminator is (Conv -> BatchNorm -> Leaky ReLU)
- The shape of the output after the last layer is (batch_size, 30, 30, 1)
- Each 30x30 patch of the output classifies a 70x70 portion of the input image (such an architecture is called a PatchGAN).
- Discriminator receives 2 inputs.
- Input image and the target image, which it should classify as real.
- Input image and the generated image (output of generator), which it should classify as fake.
- We concatenate these 2 inputs together in the code (
tf.concat([inp, tar], axis=-1))
def Discriminator():
initializer = tf.random_normal_initializer(0., 0.02)
inp = tf.keras.layers.Input(shape=[None, None, 3], name='input_image')
tar = tf.keras.layers.Input(shape=[None, None, 3], name='target_image')
x = tf.keras.layers.concatenate([inp, tar]) # (bs, 256, 256, channels*2)
down1 = downsample(64, 4, False)(x) # (bs, 128, 128, 64)
down2 = downsample(128, 4)(down1) # (bs, 64, 64, 128)
down3 = downsample(256, 4)(down2) # (bs, 32, 32, 256)
zero_pad1 = tf.keras.layers.ZeroPadding2D()(down3) # (bs, 34, 34, 256)
conv = tf.keras.layers.Conv2D(512, 4, strides=1,
kernel_initializer=initializer,
use_bias=False)(zero_pad1) # (bs, 31, 31, 512)
batchnorm1 = tf.keras.layers.BatchNormalization()(conv)
leaky_relu = tf.keras.layers.LeakyReLU()(batchnorm1)
zero_pad2 = tf.keras.layers.ZeroPadding2D()(leaky_relu) # (bs, 33, 33, 512)
last = tf.keras.layers.Conv2D(1, 4, strides=1,
kernel_initializer=initializer)(zero_pad2) # (bs, 30, 30, 1)
return tf.keras.Model(inputs=[inp, tar], outputs=last)discriminator = Discriminator()
disc_out = discriminator([inp[tf.newaxis,...], gen_output], training=False)You can find more details in Tensorflow 2.0 implementations Here
@Tensorflow2.0
