cnn.py

'''Neural style transfer with Keras.

Before running this script, download the weights for the VGG16 model at:
https://drive.google.com/file/d/0Bz7KyqmuGsilT0J5dmRCM0ROVHc/view?usp=sharing
(source: https://gist.github.com/baraldilorenzo/07d7802847aaad0a35d3)
and make sure the variable `weights_path` in this script matches the location of the file.

Run the script with:
```
python neural_style_transfer.py path_to_your_base_image.jpg path_to_your_reference.jpg prefix_for_results
```
e.g.:
```
python neural_style_transfer.py img/tuebingen.jpg img/starry_night.jpg results/my_result
```

It is preferable to run this script on GPU, for speed.
If running on CPU, prefer the TensorFlow backend (much faster).

Example result: https://twitter.com/fchollet/status/686631033085677568

# Details

Style transfer consists in generating an image
with the same "content" as a base image, but with the
"style" of a different picture (typically artistic).

This is achieved through the optimization of a loss function
that has 3 components: "style loss", "content loss",
and "total variation loss":

- The total variation loss imposes local spatial continuity between
the pixels of the combination image, giving it visual coherence.

- The style loss is where the deep learning keeps in --that one is defined
using a deep convolutional neural network. Precisely, it consists in a sum of
L2 distances between the Gram matrices of the representations of
the base image and the style reference image, extracted from
different layers of a convnet (trained on ImageNet). The general idea
is to capture color/texture information at different spatial
scales (fairly large scales --defined by the depth of the layer considered).

 - The content loss is a L2 distance between the features of the base
image (extracted from a deep layer) and the features of the combination image,
keeping the generated image close enough to the original one.

# References
    - [A Neural Algorithm of Artistic Style](http://arxiv.org/abs/1508.06576)
'''

from __future__ import print_function
from scipy.misc import imread, imresize, imsave
import numpy as np
from scipy.optimize import fmin_l_bfgs_b
import time
import os
import argparse
import h5py
import tinys3

from keras.models import Sequential
from keras.layers import Convolution2D, ZeroPadding2D, MaxPooling2D
from keras import backend as K

parser = argparse.ArgumentParser(description='Neural style transfer with Keras.')
parser.add_argument('base_image_path', metavar='base', type=str,
                    help='Path to the image to transform.')
parser.add_argument('style_reference_image_path', metavar='ref', type=str,
                    help='Path to the style reference image.')
parser.add_argument('batch_id', metavar='batch_id', type=str,
                    help='Batch id.')
parser.add_argument('frame', metavar='frame', type=str,
                    help='Frame number.')
args = parser.parse_args()
base_image_path = args.base_image_path
style_reference_image_path = args.style_reference_image_path
batch_id = args.batch_id
frame = args.frame
weights_path = 'vgg16_weights.h5'

# these are the weights of the different loss components
total_variation_weight = 1.
style_weight = 1.
content_weight = 0.025


# dimensions of the generated picture.
img_width = 400
img_height = 400
assert img_height == img_width, 'Due to the use of the Gram matrix, width and height must match.'

# util function to open, resize and format pictures into appropriate tensors
def preprocess_image(image_path):
    img = imresize(imread(image_path), (img_width, img_height))
    img = img[:, :, ::-1].astype('float64')
    img[:, :, 0] -= 103.939
    img[:, :, 1] -= 116.779
    img[:, :, 2] -= 123.68
    img = img.transpose((2, 0, 1))
    img = np.expand_dims(img, axis=0)
    return img

# util function to convert a tensor into a valid image
def deprocess_image(x):
    x = x.transpose((1, 2, 0))
    x[:, :, 0] += 103.939
    x[:, :, 1] += 116.779
    x[:, :, 2] += 123.68
    x = x[:, :, ::-1]
    x = np.clip(x, 0, 255).astype('uint8')
    return x

# get tensor representations of our images
base_image = K.variable(preprocess_image(base_image_path))
style_reference_image = K.variable(preprocess_image(style_reference_image_path))

# this will contain our generated image
combination_image = K.placeholder((1, 3, img_width, img_height))

# combine the 3 images into a single Keras tensor
input_tensor = K.concatenate([base_image,
                              style_reference_image,
                              combination_image], axis=0)

# build the VGG16 network with our 3 images as input
first_layer = ZeroPadding2D((1, 1))
first_layer.set_input(input_tensor, shape=(3, 3, img_width, img_height))

model = Sequential()
model.add(first_layer)
model.add(Convolution2D(64, 3, 3, activation='relu', name='conv1_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(64, 3, 3, activation='relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))

model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(128, 3, 3, activation='relu', name='conv2_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(128, 3, 3, activation='relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))

model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu', name='conv3_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))

model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv4_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv4_2'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))

model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv5_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))

# load the weights of the VGG16 networks
# (trained on ImageNet, won the ILSVRC competition in 2014)
# note: when there is a complete match between your model definition
# and your weight savefile, you can simply call model.load_weights(filename)
assert os.path.exists(weights_path), 'Model weights not found (see "weights_path" variable in script).'
f = h5py.File(weights_path)
for k in range(f.attrs['nb_layers']):
    if k >= len(model.layers):
        # we don't look at the last (fully-connected) layers in the savefile
        break
    g = f['layer_{}'.format(k)]
    weights = [g['param_{}'.format(p)] for p in range(g.attrs['nb_params'])]
    model.layers[k].set_weights(weights)
f.close()
print('Model loaded.')

# 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])

# compute the neural style loss
# first we need to define 4 util functions

# the gram matrix of an image tensor (feature-wise outer product)
def gram_matrix(x):
    assert K.ndim(x) == 3
    features = K.batch_flatten(x)
    gram = K.dot(features, K.transpose(features))
    return gram

# the "style loss" is designed to maintain
# the style of the reference image in the generated image.
# It is based on the gram matrices (which capture style) of
# feature maps from the style reference image
# and from the generated image
def style_loss(style, combination):
    assert K.ndim(style) == 3
    assert K.ndim(combination) == 3
    S = gram_matrix(style)
    C = gram_matrix(combination)
    channels = 3
    size = img_width * img_height
    return K.sum(K.square(S - C)) / (4. * (channels ** 2) * (size ** 2))

# an auxiliary loss function
# designed to maintain the "content" of the
# base image in the generated image
def content_loss(base, combination):
    return K.sum(K.square(combination - base))

# the 3rd loss function, total variation loss,
# designed to keep the generated image locally coherent
def total_variation_loss(x):
    assert K.ndim(x) == 4
    a = K.square(x[:, :, :img_width-1, :img_height-1] - x[:, :, 1:, :img_height-1])
    b = K.square(x[:, :, :img_width-1, :img_height-1] - x[:, :, :img_width-1, 1:])
    return K.sum(K.pow(a + b, 1.25))

# combine these loss functions into a single scalar
loss = K.variable(0.)
layer_features = outputs_dict['conv4_2']
base_image_features = layer_features[0, :, :, :]
combination_features = layer_features[2, :, :, :]
loss += content_weight * content_loss(base_image_features,
                                      combination_features)

feature_layers = ['conv1_1', 'conv2_1', 'conv3_1', 'conv4_1', 'conv5_1']
for layer_name in feature_layers:
    layer_features = outputs_dict[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(feature_layers)) * sl
loss += total_variation_weight * total_variation_loss(combination_image)

# get the gradients of the generated image wrt the loss
grads = K.gradients(loss, combination_image)

outputs = [loss]
if type(grads) in {list, tuple}:
    outputs += grads
else:
    outputs.append(grads)

f_outputs = K.function([combination_image], outputs)
def eval_loss_and_grads(x):
    x = x.reshape((1, 3, img_width, img_height))
    outs = f_outputs([x])
    loss_value = outs[0]
    if len(outs[1:]) == 1:
        grad_values = outs[1].flatten().astype('float64')
    else:
        grad_values = np.array(outs[1:]).flatten().astype('float64')
    return loss_value, grad_values

# this Evaluator class makes it possible
# to compute loss and gradients in one pass
# while retrieving them via two separate functions,
# "loss" and "grads". This is done because scipy.optimize
# requires separate functions for loss and gradients,
# but computing them separately would be inefficient.
class Evaluator(object):
    def __init__(self):
        self.loss_value = None
        self.grads_values = None

    def loss(self, x):
        assert self.loss_value is None
        loss_value, grad_values = eval_loss_and_grads(x)
        self.loss_value = loss_value
        self.grad_values = grad_values
        return self.loss_value

    def grads(self, x):
        assert self.loss_value is not None
        grad_values = np.copy(self.grad_values)
        self.loss_value = None
        self.grad_values = None
        return grad_values

evaluator = Evaluator()

# run scipy-based optimization (L-BFGS) over the pixels of the generated image
# so as to minimize the neural style loss
x = np.random.uniform(0, 255, (1, 3, img_width, img_height))
x[0, 0, :, :] -= 103.939
x[0, 1, :, :] -= 116.779
x[0, 2, :, :] -= 123.68

# Create workfolder
if not os.path.exists(batch_id):
    os.makedirs(batch_id)

for i in range(10):
    print('Start of iteration', i)
    start_time = time.time()
    x, min_val, info = fmin_l_bfgs_b(evaluator.loss, x.flatten(),
                                     fprime=evaluator.grads, maxfun=20)
    print('Current loss value:', min_val)
    # save current generated image
    img = deprocess_image(x.copy().reshape((3, img_width, img_height)))
    fname = batch_id + '/' + frame + '-%d.png' % i
    imsave(fname, img)
    end_time = time.time()
    print('Image saved as', fname)

    # Upload to S3
    f = open(fname,'rb')
    conn = tinys3.Connection("******","******",tls=True)
    conn.upload(fname, f, 'kubernets-artist')
    print('Image uploaded to s3')

    print('Iteration %d completed in %ds' % (i, end_time - start_time))