深度有趣 | 04 图像风格迁移

图像风格迁移是指，将一幅内容图的内容，和一幅或多幅风格图的风格融合在一起，从而生成一些有意思的图片

以下是将一些艺术作品的风格，迁移到一张内容图之后的效果

我们使用 TensorFlow 和 Keras 分别来实现图像风格迁移，主要用到深度学习中的卷积神经网络，即CNN

准备

安装包

pip install numpy scipy tensorflow keras
复制代码

再准备一些风格图片，和一张内容图片

原理

为了将风格图的风格和内容图的内容进行融合，所生成的图片，在内容上应当尽可能接近内容图，在风格上应当尽可能接近风格图

因此需要定义 内容损失函数 和 风格损失函数 ，经过加权后作为总的损失函数

实现步骤如下

• 随机产生一张图片
• 在每轮迭代中，根据总的损失函数，调整图片的像素值
• 经过多轮迭代，得到优化后的图片

内容损失函数

两张图片在内容上相似，不能仅仅靠简单的纯像素比较

CNN具有抽象和理解图像的能力，因此可以考虑将各个卷积层的输出作为图像的内容

以 VGG19 为例，其中包括了多个卷积层、池化层，以及最后的全连接层

这里我们使用 conv4_2 的输出作为图像的内容表示，定义内容损失函数如下

L_{content}(\vec{p},\vec{x},l)=\frac{1}{2}\sum_{i,j}{(F_{ij}^{l}-P_{ij}^{l})}^2
复制代码

风格损失函数

风格是一个很难说清楚的概念，可能是笔触、纹理、结构、布局、用色等等

这里我们使用卷积层各个特征图之间的互相关作为图像的风格，以 conv1_1 为例

Gram

Gram 矩阵的计算如下，如果有64个特征图，那么 Gram 矩阵的大小便是 64*64 ，第 i 行第 j 列的值表示第 i 个特征图和第 j 个特征图之间的互相关，用内积计算

G_{ij}^l=\sum_k{F_{ik}^l F_{jk}^l}
复制代码

风格损失函数定义如下，对多个卷积层的风格表示差异进行加权

E_l=\frac{1}{4N_l^2 M_l^2}\sum_{i,j}(G_{ij}^l-A_{ij}^l)^2

L_{style}(\vec{a},\vec{x})=\sum_{l=0}^{L}\omega_l E_l

复制代码

这里我们使用 conv1_1 、 conv2_1 、 conv3_1 、 conv4_1 、 conv5_1 五个卷积层，进行风格损失函数的计算，不同的权重会导致不同的迁移效果

总的损失函数

总的损失函数即内容损失函数和风格损失函数的加权，不同的权重会导致不同的迁移效果

L_{total}(\vec{p},\vec{a},\vec{x})=\alpha L_{content}(\vec{p},\vec{x})+\beta L_{style}(\vec{a},\vec{x})
复制代码

TensorFlow实现

加载库

# -*- coding: utf-8 -*-

import tensorflow as tf
import numpy as np
import scipy.io
import scipy.misc
import os
import time

def the_current_time():
	print(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(int(time.time()))))
复制代码

定义一些变量

CONTENT_IMG = 'content.jpg'
STYLE_IMG = 'style5.jpg'
OUTPUT_DIR = 'neural_style_transfer_tensorflow/'

if not os.path.exists(OUTPUT_DIR):
	os.mkdir(OUTPUT_DIR)

IMAGE_W = 800
IMAGE_H = 600
COLOR_C = 3

NOISE_RATIO = 0.7
BETA = 5
ALPHA = 100
VGG_MODEL = 'imagenet-vgg-verydeep-19.mat'
MEAN_VALUES = np.array([123.68, 116.779, 103.939]).reshape((1, 1, 1, 3))
复制代码

加载 VGG19 模型

def load_vgg_model(path):
	'''
	Details of the VGG19 model:
	- 0 is conv1_1 (3, 3, 3, 64)
	- 1 is relu
	- 2 is conv1_2 (3, 3, 64, 64)
	- 3 is relu    
	- 4 is maxpool
	- 5 is conv2_1 (3, 3, 64, 128)
	- 6 is relu
	- 7 is conv2_2 (3, 3, 128, 128)
	- 8 is relu
	- 9 is maxpool
	- 10 is conv3_1 (3, 3, 128, 256)
	- 11 is relu
	- 12 is conv3_2 (3, 3, 256, 256)
	- 13 is relu
	- 14 is conv3_3 (3, 3, 256, 256)
	- 15 is relu
	- 16 is conv3_4 (3, 3, 256, 256)
	- 17 is relu
	- 18 is maxpool
	- 19 is conv4_1 (3, 3, 256, 512)
	- 20 is relu
	- 21 is conv4_2 (3, 3, 512, 512)
	- 22 is relu
	- 23 is conv4_3 (3, 3, 512, 512)
	- 24 is relu
	- 25 is conv4_4 (3, 3, 512, 512)
	- 26 is relu
	- 27 is maxpool
	- 28 is conv5_1 (3, 3, 512, 512)
	- 29 is relu
	- 30 is conv5_2 (3, 3, 512, 512)
	- 31 is relu
	- 32 is conv5_3 (3, 3, 512, 512)
	- 33 is relu
	- 34 is conv5_4 (3, 3, 512, 512)
	- 35 is relu
	- 36 is maxpool
	- 37 is fullyconnected (7, 7, 512, 4096)
	- 38 is relu
	- 39 is fullyconnected (1, 1, 4096, 4096)
	- 40 is relu
	- 41 is fullyconnected (1, 1, 4096, 1000)
	- 42 is softmax
	'''
	vgg = scipy.io.loadmat(path)
	vgg_layers = vgg['layers']

	def _weights(layer, expected_layer_name):
		W = vgg_layers[0][layer][0][0][2][0][0]
		b = vgg_layers[0][layer][0][0][2][0][1]
		layer_name = vgg_layers[0][layer][0][0][0][0]
		assert layer_name == expected_layer_name
		return W, b

	def _conv2d_relu(prev_layer, layer, layer_name):
		W, b = _weights(layer, layer_name)
		W = tf.constant(W)
		b = tf.constant(np.reshape(b, (b.size)))
		return tf.nn.relu(tf.nn.conv2d(prev_layer, filter=W, strides=[1, 1, 1, 1], padding='SAME') + b)

	def _avgpool(prev_layer):
		return tf.nn.avg_pool(prev_layer, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')

	graph = {}
	graph['input']    = tf.Variable(np.zeros((1, IMAGE_H, IMAGE_W, COLOR_C)), dtype='float32')
	graph['conv1_1']  = _conv2d_relu(graph['input'], 0, 'conv1_1')
	graph['conv1_2']  = _conv2d_relu(graph['conv1_1'], 2, 'conv1_2')
	graph['avgpool1'] = _avgpool(graph['conv1_2'])
	graph['conv2_1']  = _conv2d_relu(graph['avgpool1'], 5, 'conv2_1')
	graph['conv2_2']  = _conv2d_relu(graph['conv2_1'], 7, 'conv2_2')
	graph['avgpool2'] = _avgpool(graph['conv2_2'])
	graph['conv3_1']  = _conv2d_relu(graph['avgpool2'], 10, 'conv3_1')
	graph['conv3_2']  = _conv2d_relu(graph['conv3_1'], 12, 'conv3_2')
	graph['conv3_3']  = _conv2d_relu(graph['conv3_2'], 14, 'conv3_3')
	graph['conv3_4']  = _conv2d_relu(graph['conv3_3'], 16, 'conv3_4')
	graph['avgpool3'] = _avgpool(graph['conv3_4'])
	graph['conv4_1']  = _conv2d_relu(graph['avgpool3'], 19, 'conv4_1')
	graph['conv4_2']  = _conv2d_relu(graph['conv4_1'], 21, 'conv4_2')
	graph['conv4_3']  = _conv2d_relu(graph['conv4_2'], 23, 'conv4_3')
	graph['conv4_4']  = _conv2d_relu(graph['conv4_3'], 25, 'conv4_4')
	graph['avgpool4'] = _avgpool(graph['conv4_4'])
	graph['conv5_1']  = _conv2d_relu(graph['avgpool4'], 28, 'conv5_1')
	graph['conv5_2']  = _conv2d_relu(graph['conv5_1'], 30, 'conv5_2')
	graph['conv5_3']  = _conv2d_relu(graph['conv5_2'], 32, 'conv5_3')
	graph['conv5_4']  = _conv2d_relu(graph['conv5_3'], 34, 'conv5_4')
	graph['avgpool5'] = _avgpool(graph['conv5_4'])
	return graph
复制代码

内容损失函数

def content_loss_func(sess, model):
	def _content_loss(p, x):
		N = p.shape[3]
		M = p.shape[1] * p.shape[2]
		return (1 / (4 * N * M)) * tf.reduce_sum(tf.pow(x - p, 2))
	return _content_loss(sess.run(model['conv4_2']), model['conv4_2'])
复制代码

风格损失函数

STYLE_LAYERS = [('conv1_1', 0.5), ('conv2_1', 1.0), ('conv3_1', 1.5), ('conv4_1', 3.0), ('conv5_1', 4.0)]

def style_loss_func(sess, model):
	def _gram_matrix(F, N, M):
		Ft = tf.reshape(F, (M, N))
		return tf.matmul(tf.transpose(Ft), Ft)

	def _style_loss(a, x):
		N = a.shape[3]
		M = a.shape[1] * a.shape[2]
		A = _gram_matrix(a, N, M)
		G = _gram_matrix(x, N, M)
		return (1 / (4 * N ** 2 * M ** 2)) * tf.reduce_sum(tf.pow(G - A, 2))

	return sum([_style_loss(sess.run(model[layer_name]), model[layer_name]) * w for layer_name, w in STYLE_LAYERS])
复制代码

随机产生一张初始图片

def generate_noise_image(content_image, noise_ratio=NOISE_RATIO):
	noise_image = np.random.uniform(-20, 20, (1, IMAGE_H, IMAGE_W, COLOR_C)).astype('float32')
	input_image = noise_image * noise_ratio + content_image * (1 - noise_ratio)
	return input_image
复制代码

加载图片

def load_image(path):
	image = scipy.misc.imread(path)
	image = scipy.misc.imresize(image, (IMAGE_H, IMAGE_W))
	image = np.reshape(image, ((1, ) + image.shape))
	image = image - MEAN_VALUES
	return image
复制代码

保存图片

def save_image(path, image):
	image = image + MEAN_VALUES
	image = image[0]
	image = np.clip(image, 0, 255).astype('uint8')
	scipy.misc.imsave(path, image)
复制代码

调用以上函数并训练模型

the_current_time()

with tf.Session() as sess:
	content_image = load_image(CONTENT_IMG)
	style_image = load_image(STYLE_IMG)
	model = load_vgg_model(VGG_MODEL)

	input_image = generate_noise_image(content_image)
	sess.run(tf.global_variables_initializer())

	sess.run(model['input'].assign(content_image))
	content_loss = content_loss_func(sess, model)

	sess.run(model['input'].assign(style_image))
	style_loss = style_loss_func(sess, model)

	total_loss = BETA * content_loss + ALPHA * style_loss
	optimizer = tf.train.AdamOptimizer(2.0)
	train = optimizer.minimize(total_loss)

	sess.run(tf.global_variables_initializer())
	sess.run(model['input'].assign(input_image))

	ITERATIONS = 2000
	for i in range(ITERATIONS):
		sess.run(train)
		if i % 100 == 0:
			output_image = sess.run(model['input'])
			the_current_time()
			print('Iteration %d' % i)
			print('Cost: ', sess.run(total_loss))

			save_image(os.path.join(OUTPUT_DIR, 'output_%d.jpg' % i), output_image)
复制代码

在GPU上跑，花了5分钟左右，2000轮迭代后是这个样子

对比原图

Keras实现

Keras官方提供了图像风格迁移的例子

github.com/fchollet/ke…

代码里引入了一个 total variation loss ，翻译为全变差正则，据说可以让生成的图像更平滑

conv5_2

代码使用方法如下

python neural_style_transfer.py path_to_your_base_image.jpg path_to_your_reference.jpg prefix_for_results
复制代码

--iter
--content_weight
--style_weight
--tv_weight

新建文件夹 neural_style_transfer_keras

python main_keras.py content.jpg style5.jpg neural_style_transfer_keras/output
复制代码

生成的图片长这样，10次迭代，花了1分钟左右