TensorFlowでスタイル変換して遊んだ【2022】

# ライブラリのインポート
from tqdm import tqdm
import numpy as np
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras.applications import vgg19

base_image_path = keras.utils.get_file("paris.jpg", "https://i.imgur.com/F28w3Ac.jpg")
style_reference_image_path = keras.utils.get_file(
    "starry_night.jpg", "https://i.imgur.com/9ooB60I.jpg"
)
result_prefix = "paris_generated"

# 異なる三つの重みを設定
total_variation_weight = 1e-6
style_weight = 1e-6
content_weight = 2.5e-8

# 生成した画像の次元(サイズ)を設定
width, height = keras.preprocessing.image.load_img(base_image_path).size
img_nrows = 400
img_ncols = int(width * img_nrows / height)

# ベース（コンテンツ）画像とスタイル参照画像を見てみる

from IPython.display import Image, display

display(Image(base_image_path))
display(Image(style_reference_image_path))

# 画像処理のふたつの関数を定義
# 画像の前処理(画像→テンソル)
def preprocess_image(image_path):
    # 画像を開いてサイズを変更し、適切なテンソルにフォーマットするためのUtil関数
    img = keras.preprocessing.image.load_img(
        image_path, target_size=(img_nrows, img_ncols)
    )
    img = keras.preprocessing.image.img_to_array(img)
    img = np.expand_dims(img, axis=0)
    img = vgg19.preprocess_input(img)
    return tf.convert_to_tensor(img)

# テンソルの後処理(テンソル→画像)
def deprocess_image(x):
    # テンソルを有効な画像に変換するためのUtil関数
    x = x.reshape((img_nrows, img_ncols, 3))
    # Remove zero-center by mean pixel
    x[:, :, 0] += 103.939
    x[:, :, 1] += 116.779
    x[:, :, 2] += 123.68
    # 'BGR'->'RGB'
    x = x[:, :, ::-1]
    x = np.clip(x, 0, 255).astype("uint8")
    return x

# グラム行列の定義
def gram_matrix(x):
    x = tf.transpose(x, (2, 0, 1))
    features = tf.reshape(x, (tf.shape(x)[0], -1))
    gram = tf.matmul(features, tf.transpose(features))
    return gram

# スタイル損失は、生成画像において参照画像のスタイルを維持するためのものである。
# グラム行列（スタイルを表す）に基づくものである特徴マップを生成する。

# スタイル損失の定義
def style_loss(style, combination):
    S = gram_matrix(style)
    C = gram_matrix(combination)
    channels = 3
    size = img_nrows * img_ncols
    return tf.reduce_sum(tf.square(S - C)) / (4.0 * (channels**2) * (size**2))

# An auxiliary loss function
# designed to maintain the "content" of the
# base image in the generated image

# コンテンツ損失の定義
def content_loss(base, combination):
    return tf.reduce_sum(tf.square(combination - base))


# The 3rd loss function, total variation loss,
# designed to keep the generated image locally coherent

# 全変動損失
def total_variation_loss(x):
    a = tf.square(
        x[:, : img_nrows - 1, : img_ncols - 1, :] - x[:, 1:, : img_ncols - 1, :]
    )
    b = tf.square(
        x[:, : img_nrows - 1, : img_ncols - 1, :] - x[:, : img_nrows - 1, 1:, :]
    )
    return tf.reduce_sum(tf.pow(a + b, 1.25))

# Build a VGG19 model loaded with pre-trained ImageNet weights
model = vgg19.VGG19(weights="imagenet", include_top=False)

# Get the symbolic outputs of each "key" layer (we gave them unique names).
outputs_dict = dict([(layer.name, layer.output) for layer in model.layers])

# vgg16の各レイヤーの活性化値を返すモデルを設定する。
feature_extractor = keras.Model(inputs=model.inputs, outputs=outputs_dict)

# List of layers to use for the style loss.
style_layer_names = [
    "block1_conv1",
    "block2_conv1",
    "block3_conv1",
    "block4_conv1",
    "block5_conv1",
]
# The layer to use for the content loss.
content_layer_name = "block5_conv2"

# 損失の計算関数
def compute_loss(combination_image, base_image, style_reference_image):
    input_tensor = tf.concat(
        [base_image, style_reference_image, combination_image], axis=0
    )
    features = feature_extractor(input_tensor)

    # Initialize the loss
    loss = tf.zeros(shape=())

    # Add content loss
    layer_features = features[content_layer_name]
    base_image_features = layer_features[0, :, :, :]
    combination_features = layer_features[2, :, :, :]
    loss = loss + content_weight * content_loss(
        base_image_features, combination_features
    )
    # Add style loss
    for layer_name in style_layer_names:
        layer_features = features[layer_name]
        style_reference_features = layer_features[1, :, :, :]
        combination_features = layer_features[2, :, :, :]
        sl = style_loss(style_reference_features, combination_features)
        loss += (style_weight / len(style_layer_names)) * sl

    # Add total variation loss
    loss += total_variation_weight * total_variation_loss(combination_image)
    return loss

@tf.function
def compute_loss_and_grads(combination_image, base_image, style_reference_image):
    with tf.GradientTape() as tape:
        loss = compute_loss(combination_image, base_image, style_reference_image)
    grads = tape.gradient(loss, combination_image)
    return loss, grads

# 最適化関数の定義
optimizer = keras.optimizers.SGD(
    keras.optimizers.schedules.ExponentialDecay(
        initial_learning_rate=100.0, decay_steps=100, decay_rate=0.96
    )
)

base_image = preprocess_image(base_image_path)
style_reference_image = preprocess_image(style_reference_image_path)
combination_image = tf.Variable(preprocess_image(base_image_path))

iterations = 4000

# トレーニング実行
for i in tqdm(range(1, iterations + 1)):
    loss, grads = compute_loss_and_grads(
        combination_image, base_image, style_reference_image
    )
    optimizer.apply_gradients([(grads, combination_image)])
    if i % 100 == 0:
        print("Iteration %d: loss=%.2f" % (i, loss))
        img = deprocess_image(combination_image.numpy())
        fname = result_prefix + "_at_iteration_%d.png" % i
        keras.preprocessing.image.save_img(fname, img)

display(Image(result_prefix + "_at_iteration_4000.png"))

# h5形式で保存すると、optimizerの重みが更新されない
model.save('style-4000')

# ライブラリのインポート
from tqdm import tqdm
import numpy as np
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras.applications import vgg19

# モデル読み込み
model = tf.keras.models.load_model('style-4000')

# Get the symbolic outputs of each "key" layer (we gave them unique names).
outputs_dict = dict([(layer.name, layer.output) for layer in model.layers])

# vgg16の各レイヤーの活性化値を返すモデルを設定する。
feature_extractor = keras.Model(inputs=model.inputs, outputs=outputs_dict)

# 損失の計算関数
def compute_loss(combination_image, base_image, style_reference_image):
    input_tensor = tf.concat(
        [base_image, style_reference_image, combination_image], axis=0
    )
    features = feature_extractor(input_tensor)

    # Initialize the loss
    loss = tf.zeros(shape=())

    # Add content loss
    layer_features = features[content_layer_name]
    base_image_features = layer_features[0, :, :, :]
    combination_features = layer_features[2, :, :, :]
    loss = loss + content_weight * content_loss(
        base_image_features, combination_features
    )
    # Add style loss
    for layer_name in style_layer_names:
        layer_features = features[layer_name]
        style_reference_features = layer_features[1, :, :, :]
        combination_features = layer_features[2, :, :, :]
        sl = style_loss(style_reference_features, combination_features)
        loss += (style_weight / len(style_layer_names)) * sl

    # Add total variation loss
    loss += total_variation_weight * total_variation_loss(combination_image)
    return loss

@tf.function
def compute_loss_and_grads(combination_image, base_image, style_reference_image):
    with tf.GradientTape() as tape:
        loss = compute_loss(combination_image, base_image, style_reference_image)
    grads = tape.gradient(loss, combination_image)
    return loss, grads

base_image_path = 'style_test.jpg'
style_reference_image_path = keras.utils.get_file(
    "starry_night.jpg", "https://i.imgur.com/9ooB60I.jpg"
)

from IPython.display import Image, display

display(Image(base_image_path))
display(Image(style_reference_image_path))

loss, grads = compute_loss_and_grads(
    combination_image, base_image, style_reference_image
)
optimizer.apply_gradients([(grads, combination_image)])
img = deprocess_image(combination_image.numpy())
fname = "style-4000-1.png"
keras.preprocessing.image.save_img(fname, img)
        

display(Image(fname))