app.py

import os
import tensorflow as tf
# Load compressed models from tensorflow_hub
os.environ['TFHUB_MODEL_LOAD_FORMAT'] = 'COMPRESSED'
import IPython.display as display

import matplotlib.pyplot as plt
import matplotlib as mpl
mpl.rcParams['figure.figsize'] = (12, 12)
mpl.rcParams['axes.grid'] = False

import numpy as np
import PIL.Image

def tensor_to_image(tensor):
  tensor = tensor*255
  tensor = np.array(tensor, dtype=np.uint8)
  if np.ndim(tensor)>3:
    assert tensor.shape[0] == 1
    tensor = tensor[0]
  return PIL.Image.fromarray(tensor)

def load_img(path_to_img):
  max_dim = 1024
  img = tf.io.read_file(path_to_img)
  img = tf.image.decode_image(img, channels=3)
  img = tf.image.convert_image_dtype(img, tf.float32)

  shape = tf.cast(tf.shape(img)[:-1], tf.float32)
  long_dim = max(shape)
  scale = max_dim / long_dim

  new_shape = tf.cast(shape * scale, tf.int32)

  img = tf.image.resize(img, new_shape)
  img = img[tf.newaxis, :]
  return img

def imshow(image, title=None):
  if len(image.shape) > 3:
    image = tf.squeeze(image, axis=0)

  plt.imshow(image)
  if title:
    plt.title(title)

content_layers = ['block5_conv2'] 

style_layers = ['block1_conv1',
                'block2_conv1',
                'block3_conv1', 
                'block4_conv1', 
                'block5_conv1']

num_content_layers = len(content_layers)
num_style_layers = len(style_layers)

def vgg_layers(layer_names):
  """ Creates a vgg model that returns a list of intermediate output values."""
  # Load our model. Load pretrained VGG, trained on imagenet data
  vgg = tf.keras.applications.VGG19(include_top=False, weights='imagenet')
  vgg.trainable = False
  
  outputs = [vgg.get_layer(name).output for name in layer_names]

  model = tf.keras.Model([vgg.input], outputs)
  return model

def gram_matrix(input_tensor):
  result = tf.linalg.einsum('bijc,bijd->bcd', input_tensor, input_tensor)
  input_shape = tf.shape(input_tensor)
  num_locations = tf.cast(input_shape[1]*input_shape[2], tf.float32)
  return result/(num_locations)

class StyleContentModel(tf.keras.models.Model):
  def __init__(self, style_layers, content_layers):
    super(StyleContentModel, self).__init__()
    self.vgg = vgg_layers(style_layers + content_layers)
    self.style_layers = style_layers
    self.content_layers = content_layers
    self.num_style_layers = len(style_layers)
    self.vgg.trainable = False

  def call(self, inputs):
    "Expects float input in [0,1]"
    inputs = inputs*255.0
    preprocessed_input = tf.keras.applications.vgg19.preprocess_input(inputs)
    outputs = self.vgg(preprocessed_input)
    style_outputs, content_outputs = (outputs[:self.num_style_layers],
                                      outputs[self.num_style_layers:])

    style_outputs = [gram_matrix(style_output)
                     for style_output in style_outputs]

    content_dict = {content_name: value
                    for content_name, value
                    in zip(self.content_layers, content_outputs)}

    style_dict = {style_name: value
                  for style_name, value
                  in zip(self.style_layers, style_outputs)}

    return {'content': content_dict, 'style': style_dict}

extractor = StyleContentModel(style_layers, content_layers)

def clip_0_1(image):
  return tf.clip_by_value(image, clip_value_min=0.0, clip_value_max=1.0)

def high_pass_x_y(image):
  x_var = image[:, :, 1:, :] - image[:, :, :-1, :]
  y_var = image[:, 1:, :, :] - image[:, :-1, :, :]

  return x_var, y_var

def total_variation_loss(image):
  x_deltas, y_deltas = high_pass_x_y(image)
  return tf.reduce_sum(tf.abs(x_deltas)) + tf.reduce_sum(tf.abs(y_deltas))

opt = tf.optimizers.Adam(learning_rate=0.02, beta_1=0.99, epsilon=1e-1)

style_weight=1e-2
content_weight=1e4
total_variation_weight=30

epochs = 10
steps_per_epoch = 50

def transfer_style(content_path,style_path,transfer_mode,steps_per_epoch=100,style_weight=1e-2,content_weight=1e4,total_variation_weight=30):
  try:

    content_image = load_img(content_path)
    style_image = load_img(style_path)
    if transfer_mode == "Fast_transfer":
      res = transfer_style_fast(content_image,style_image)
    else:
      res = transfer_style_custom(content_image,style_image,int(steps_per_epoch),style_weight,content_weight,total_variation_weight)
    res = tensor_to_image(res)
  except Exception as ex:
    raise Exception(ex)
  return res

def transfer_style_fast(content_image,style_image):
  import tensorflow_hub as hub
  hub_model = hub.load('https://tfhub.dev/google/magenta/arbitrary-image-stylization-v1-256/2')
  return hub_model(tf.constant(content_image), tf.constant(style_image))[0]

def transfer_style_custom(content_image,style_image,steps_per_epoch=100,style_weight=1e-2,content_weight=1e4,total_variation_weight=30):

  def style_content_loss(outputs):
    style_outputs = outputs['style']
    content_outputs = outputs['content']
    style_loss = tf.add_n([tf.reduce_mean((style_outputs[name]-style_targets[name])**2) 
                           for name in style_outputs.keys()])
    style_loss *= style_weight / num_style_layers

    content_loss = tf.add_n([tf.reduce_mean((content_outputs[name]-content_targets[name])**2) 
                             for name in content_outputs.keys()])
    content_loss *= content_weight / num_content_layers
    loss = style_loss + content_loss
    return loss

  @tf.function()
  def train_step(image):
    with tf.GradientTape() as tape:
      outputs = extractor(image)
      loss = style_content_loss(outputs)
      loss += total_variation_weight*tf.image.total_variation(image)

    grad = tape.gradient(loss, image)
    opt.apply_gradients([(grad, image)])
    image.assign(clip_0_1(image))
  try:
    style_targets = extractor(style_image)['style']
    content_targets = extractor(content_image)['content']
    image = tf.Variable(content_image)

    step = 0
    for n in range(epochs):
      for m in range(steps_per_epoch):
        step += 1
        train_step(image)
  except Exception as ex:
    raise Exception(ex)

  return image

import gradio as gr

inputs = [
          gr.inputs.Image(type="filepath"),
          gr.inputs.Image(type="filepath"),
          gr.inputs.Radio(["Fast_transfer","Custom_transfer"]),
          gr.inputs.Slider(1,100,default=30,step=1),
          gr.inputs.Number(1e-2),
          gr.inputs.Number(1e4),
          gr.inputs.Number(30)
          ]

iface = gr.Interface(
    fn=transfer_style,
    inputs=inputs,
    examples=[["NST/etsii.jpg","NST/data/style_2.jpg","Fast_transfer",30,1e-2,1e4,30],
              ["NST/data/content_9.jpg","NST/ola.png","Fast_transfer",30,1e-2,1e4,30],
              ["NST/sailboat_cropped.jpg","NST/sketch_cropped.png","Fast_transfer",30,1e-2,1e4,30],
              ["NST/armadillo.jpg","NST/data/style_3.jpg","Fast_transfer",30,1e-2,1e4,30],
              ["NST/gato.jpg","NST/data/style_4.jpg","Fast_transfer",30,1e-2,1e4,30],
              ],
    outputs="image").launch(debug=True)