-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path1024_example_percept_2.py
264 lines (210 loc) · 7.99 KB
/
1024_example_percept_2.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
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
'''
Get latent code of targte images
using perceptual loss
with pretrained network pickle. [1024x1024]
latent code: (17,32)
'''
import argparse
import math
import os
import sys
import pickle
import torch
from torch import optim
from torch.nn import functional as F
from torchvision import transforms
from PIL import Image
from tqdm import tqdm
import scipy.io as sio
import numpy as np
import csv
import misc
from misc import crop_max_rectangle as crop
import lpips
import loader
# from model import Generator
import csv
def noise_regularize(noises):
loss = 0
for noise in noises:
size = noise.shape[2]
while True:
loss = (
loss
+ (noise * torch.roll(noise, shifts=1, dims=3)).mean().pow(2)
+ (noise * torch.roll(noise, shifts=1, dims=2)).mean().pow(2)
)
if size <= 8:
break
noise = noise.reshape([-1, 1, size // 2, 2, size // 2, 2])
noise = noise.mean([3, 5])
size //= 2
return loss
def noise_normalize_(noises):
for noise in noises:
mean = noise.mean()
std = noise.std()
noise.data.add_(-mean).div_(std)
def get_lr(t, initial_lr, rampdown=0.25, rampup=0.05):
lr_ramp = min(1, (1 - t) / rampdown)
lr_ramp = 0.5 - 0.5 * math.cos(lr_ramp * math.pi)
lr_ramp = lr_ramp * min(1, t / rampup)
return initial_lr * lr_ramp
def latent_noise(latent, strength):
noise = torch.randn_like(latent) * strength
return latent + noise
def make_image(tensor):
return (
tensor.detach()
.clamp_(min=-1, max=1)
.add(1)
.div_(2)
.mul(255)
.type(torch.uint8)
.permute(0, 2, 3, 1)
.to("cpu")
.numpy()
)
def image_transform(file_path):
resize = 1024
transform = transforms.Compose(
[
transforms.Resize(resize),
transforms.CenterCrop(resize),
transforms.ToTensor(),
transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5]),
# mean = [0.485, 0.456, 0.406], std = [0.229, 0.224, 0.225]
]
)
imgs = []
I = Image.open(file_path)
I1 = I.convert("RGB")
img = transform(I1)
imgs.append(img)
imgs = torch.stack(imgs, 0).to(device)
return imgs
def image_transform2(args, file_path, size):
resize = min(args.size, size)
transform = transforms.Compose(
[
transforms.Resize(resize),
transforms.CenterCrop(resize),
transforms.ToTensor(),
transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5]),
]
)
imgs = []
img = transform(Image.open(file_path).convert("RGB").resize((size,size),Image.BILINEAR))
imgs.append(img)
imgs = torch.stack(imgs, 0).to(device)
return imgs
def projection(args, path_img1, percept, G, latent_mean, latent_std):
imgs = image_transform(path_img1)
# latent_in = latent_mean.detach().clone().unsqueeze(0).repeat(imgs.shape[0], 1, 1)
latent_in = latent_mean.detach().clone().unsqueeze(0).repeat(imgs.shape[0], 1)
# latent_in = latent_mean[None, :]
# log_size = int(math.log(args.size, 2))
# n_latent = log_size * 2 - 2
n_latent = 18
# latent_in = latent_in.unsqueeze(1).repeat(1, n_latent, 1, 1)
latent_in = latent_in.unsqueeze(1).repeat(1, n_latent, 1)
latent_in.requires_grad = True
optimizer = optim.Adam([latent_in], lr=args.lr)
loss_list = []
pbar = tqdm(range(args.step))
latent_path = []
min_loss = 1.0
for i in pbar:
t = i / args.step
lr = get_lr(t, args.lr)
# print(lr)
optimizer.param_groups[0]["lr"] = lr
noise_strength = latent_std * args.noise * max(0, 1 - t / args.noise_ramp) ** 2
latent_n = latent_noise(latent_in, noise_strength.item())
# remove: (1,18,544) --> (1,544)
# im_latent = latent_n[:,2]
# im_latent = im_latent.reshape([17,32])
# im_latent = im_latent[None, :]
# latent, average on 2nd aixs: (1,18, 544) --> (1, 544)
# im_latent = latent_n.reshape([18, 544])
im_latent = torch.mean(latent_n, 1)
im_latent = im_latent.reshape([1, 17, 32])
# im_latent = im_latent[None, :]
img_gen_raw = G(im_latent, args.truncation_psi)[0].cpu().detach().numpy()
# print(img_gen_raw)
batch, channel, height, width = img_gen_raw.shape
# if height > 256:
# factor = height // 256
# img_gen = img_gen.reshape(
# batch, channel, height // factor, factor, width // factor, factor
# )
# img_gen = img_gen.mean([3, 5])
img_gen = torch.from_numpy(img_gen_raw)
p_loss = percept(img_gen, imgs).sum()
# loss_list.append(np.ndarray(p_loss.detach().cpu().numpy()))
loss = p_loss
optimizer.zero_grad() # ?
loss.backward()
optimizer.step()
num_loss = p_loss.detach().cpu().numpy()
if num_loss < min_loss:
min_loss = num_loss
latent_path.append(im_latent.detach().clone())
# Save the image
output_dir = 'images/2_frgc_data/frgc_exp_1024/vgg'
if os.path.exists(output_dir) is False:
os.makedirs(output_dir)
pattern = "{}/sample_{{:06d}}_{{:04f}}.png".format(output_dir)
dst = crop(misc.to_pil(img_gen_raw[0]), args.ratio).save(pattern.format(i, min_loss))
pbar.set_description(
(
f"perceptual: {p_loss.item():.4f};"
f" min_loss: {min_loss.item():.4f}; lr: {lr:.4f}"
)
)
return latent_path[-1]
if __name__ == "__main__":
os.environ["CUDA_VISIBLE_DEVICES"] = "3"
device = torch.device("cuda")
parser = argparse.ArgumentParser()
# parser.add_argument("--ckpt", type=str, default='stylegan2-ffhq-config-f.pt')
# parser.add_argument("--path_to_morph", type=str, default=dst_path_morph)
# parser.add_argument("--path_to_latent", type=str, default=dst_path_latent)
parser.add_argument("--size", type=int, default=1024)
parser.add_argument("--batch_size", type=int, default=8)
parser.add_argument("--lr_rampup", type=float, default=0.05)
parser.add_argument("--lr_rampdown", type=float, default=0.25)
parser.add_argument("--lr", type=float, default=0.01)
parser.add_argument("--noise", type=float, default=0.05)
parser.add_argument("--noise_ramp", type=float, default=0.75)
parser.add_argument("--step", type=int, default=5000) # cal W
parser.add_argument("--ratio", type=float, default=1.0)
parser.add_argument("--truncation_psi", type=float, default=0.7)
parser.add_argument("--noise_regularize", type=float, default=1e5)
parser.add_argument("--w_plus", action="store_true")
args = parser.parse_args()
# Load pre-trained network
model = 'models/ffhq-snapshot-1024.pkl'
print("Loading networks...")
G = loader.load_network(model)["Gs"].to(device)
n_mean_latent = 10000
with torch.no_grad():
# Sample latent vector
# z = torch.randn([1, *G.input_shape[1:]], device=device)
# noise_sample = torch.randn(n_mean_latent, *G.input_shape[1:], device=device)
shape = G.input_shape[1:]
len = int(shape[0])*int(shape[1])
noise_sample = torch.randn(n_mean_latent,len, device=device)
# latent_out = g_ema.style(noise_sample)
latent_mean = noise_sample.mean(0)
latent_std = ((noise_sample - latent_mean).pow(2).sum() / n_mean_latent) ** 0.5
percept = lpips.PerceptualLoss(
model="net-lin", net="vgg",
use_gpu=True #device.startswith("cuda")
) #lr=0.01,beta1=0.5,version='0.1',
path_img1 = 'images/2_frgc_data/frgc_exp_1024/04827d02.png'
w1 = projection(args, path_img1, percept, G, latent_mean, latent_std)
# # Generate an image
imgs = G(w1, args.truncation_psi)[0].cpu().numpy()
dst_img = 'images/2_frgc_data/frgc_exp_1024/04827d02_latent2.png'
img = crop(misc.to_pil(imgs[0]), args.ratio).save(dst_img)