-
Notifications
You must be signed in to change notification settings - Fork 1
/
styleTransfer.py
197 lines (160 loc) · 6.82 KB
/
styleTransfer.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
import torch
import numpy as np
from PIL import Image
import torch.optim as optim
import matplotlib.pyplot as plt
from torchvision import transforms, models
###############
## Functions ##
###############
def load_image(img_path, max_size=400, shape=None):
''' Load in and transform an image, making sure the image
is <= 400 pixels in the x-y dims.'''
image = Image.open(img_path).convert('RGB')
# large images will slow down processing
if max(image.size) > max_size:
size = max_size
else:
size = max(image.size)
if shape is not None:
size = shape
in_transform = transforms.Compose([
transforms.Resize(size),
transforms.ToTensor(),
transforms.Normalize((0.485, 0.456, 0.406),
(0.229, 0.224, 0.225))])
# discard the transparent, alpha channel (that's the :3) and add the batch dimension
image = in_transform(image)[:3,:,:].unsqueeze(0)
return image
# helper function for un-normalizing an image
# and converting it from a Tensor image to a NumPy image for display
def im_convert(tensor):
""" Display a tensor as an image. """
image = tensor.to("cpu").clone().detach()
image = image.numpy().squeeze()
image = image.transpose(1,2,0)
image = image * np.array((0.229, 0.224, 0.225)) + np.array((0.485, 0.456, 0.406))
image = image.clip(0, 1)
return image
def gram_matrix(tensor):
""" Calculate the Gram Matrix of a given tensor
Gram Matrix: https://en.wikipedia.org/wiki/Gramian_matrix
"""
# get the batch_size, depth, height, and width of the Tensor
_, d, h, w = tensor.size()
# reshape so we're multiplying the features for each channel
tensor = tensor.view(d, h * w)
# calculate the gram matrix
gram = torch.mm(tensor, tensor.t())
return gram
def get_features(image, model, layers=None):
""" Run an image forward through a model and get the features for
a set of layers. Default layers are for VGGNet matching Gatys et al (2016)
"""
## TODO: Complete mapping layer names of PyTorch's VGGNet to names from the paper
## Need the layers for the content and style representations of an image
if layers is None:
layers = {'0': 'conv1_1',
'5': 'conv2_1',
'10': 'conv3_1',
'19': 'conv4_1',
'21': 'conv4_2', ## content representation
'28': 'conv5_1'}
features = {}
x = image
# model._modules is a dictionary holding each module in the model
for name, layer in model._modules.items():
x = layer(x)
if name in layers:
features[layers[name]] = x
return features
##############################
## Load in VGG19 (features) ##
##############################
# get the "features" portion of VGG19 (we will not need the "classifier" portion)
vgg = models.vgg19(pretrained=True).features
# freeze all VGG parameters since we're only optimizing the target image
for param in vgg.parameters():
param.requires_grad_(False)
# move the model to GPU, if available
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
vgg.to(device)
######################################
## Load in Content and Style Images ##
######################################
# load in content and style image
content = load_image(content).to(device)
# Resize style to match content, makes code easier
style = load_image(style, shape=content.shape[-2:]).to(device)
# display the images
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(20, 10))
# content and style ims side-by-side
ax1.imshow(im_convert(content))
ax2.imshow(im_convert(style))
##################
## VGG19 Layers ##
##################
# print out VGG19 structure so you can see the names of various layers
print(vgg)
#############################
## Putting it all Together ##
#############################
# get content and style features only once before training
content_features = get_features(content, vgg)
style_features = get_features(style, vgg)
# calculate the gram matrices for each layer of our style representation
style_grams = {layer: gram_matrix(style_features[layer]) for layer in style_features}
# create a third "target" image and prep it for change
# it is a good idea to start of with the target as a copy of our *content* image
# then iteratively change its style
target = content.clone().requires_grad_(True).to(device)
######################
## Loss and Weights ##
######################
# weights for each style layer
# weighting earlier layers more will result in *larger* style artifacts
# notice we are excluding `conv4_2` our content representation
style_weights = {'conv1_1': 1.,
'conv2_1': 0.75,
'conv3_1': 0.2,
'conv4_1': 0.2,
'conv5_1': 0.2}
content_weight = 1 # alpha
style_weight = 1e6 # beta
##############################################
## Updating the Target & Calculating Losses ##
##############################################
# for displaying the target image, intermittently
show_every = 400
# iteration hyperparameters
optimizer = optim.Adam([target], lr=0.003)
steps = 2000 # decide how many iterations to update your image (5000)
for ii in range(1, steps+1):
print(ii)
target_features = get_features(target, vgg) # get the features from your target image
content_loss = torch.mean((target_features['conv4_2'] - content_features['conv4_2'])**2) # the content loss
style_loss = 0 # initialize the style loss to 0
for layer in style_weights: # then add to it for each layer's gram matrix loss
target_feature = target_features[layer] # get the "target" style representation for the layer
target_gram = gram_matrix(target_feature)
_, d, h, w = target_feature.shape
style_gram = style_grams[layer] # get the "style" style representation
layer_style_loss = style_weights[layer] * torch.mean((target_gram - style_gram)**2) # the style loss for one layer, weighted appropriately
style_loss += layer_style_loss / (d * h * w) # add to the style loss
total_loss = content_weight * content_loss + style_weight * style_loss # calculate the *total* loss
# update your target image
optimizer.zero_grad()
total_loss.backward()
optimizer.step()
# display intermediate images and print the loss
if ii % show_every == 0:
print('Total loss: ', total_loss.item())
plt.imshow(im_convert(target))
plt.show()
##############################
## Display the Target Image ##
##############################
# display content and final, target image
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(20, 10))
ax1.imshow(im_convert(content))
ax2.imshow(im_convert(target))