-
Notifications
You must be signed in to change notification settings - Fork 24
/
Copy pathGFPGAN.py
263 lines (223 loc) · 10.3 KB
/
GFPGAN.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
import cv2
import os
import torch
from basicsr.utils import img2tensor
from facexlib.utils.face_restoration_helper import FaceRestoreHelper
from torchvision.transforms.functional import normalize
import time
from gfpgan.gfpganv1_clean_arch import GFPGANv1Clean
import time
import numpy as np
import torch.nn.functional as F
def tensor2img(tensor, rgb2bgr=True, out_type=np.uint8, min_max=(0, 1)):
"""Convert torch Tensors into image numpy arrays.
After clamping to [min, max], values will be normalized to [0, 1].
Args:
tensor (Tensor or list[Tensor]): Accept shapes:
1) 4D mini-batch Tensor of shape (B x 3/1 x H x W);
2) 3D Tensor of shape (3/1 x H x W);
3) 2D Tensor of shape (H x W).
Tensor channel should be in RGB order.
rgb2bgr (bool): Whether to change rgb to bgr.
out_type (numpy type): output types. If ``np.uint8``, transform outputs
to uint8 type with range [0, 255]; otherwise, float type with
range [0, 1]. Default: ``np.uint8``.
min_max (tuple[int]): min and max values for clamp.
Returns:
(Tensor or list): 3D ndarray of shape (H x W x C) OR 2D ndarray of
shape (H x W). The channel order is BGR.
"""
if not (torch.is_tensor(tensor) or (isinstance(tensor, list) and all(torch.is_tensor(t) for t in tensor))):
raise TypeError(f'tensor or list of tensors expected, got {type(tensor)}')
result = []
_tensor = tensor
import time
start = time.time()
_tensor = _tensor.squeeze(0).float().detach().clamp_(*min_max)
end = time.time()
_tensor = (_tensor - min_max[0]) / (min_max[1] - min_max[0])
_tensor = (_tensor.permute(1, 2, 0))
img_np = (_tensor * 255.0).round().cpu().numpy()[:, :, ::-1]
img_np = img_np.astype(out_type)
result.append(img_np)
if len(result) == 1:
result = result[0]
end = time.time()
return result
class GFPGANer():
"""Helper for restoration with GFPGAN.
It will detect and crop faces, and then resize the faces to 512x512.
GFPGAN is used to restored the resized faces.
The background is upsampled with the bg_upsampler.
Finally, the faces will be pasted back to the upsample background image.
Args:
model_path (str): The path to the GFPGAN model. It can be urls (will first download it automatically).
upscale (float): The upscale of the final output. Default: 2.
arch (str): The GFPGAN architecture. Option: clean | original. Default: clean.
channel_multiplier (int): Channel multiplier for large networks of StyleGAN2. Default: 2.
bg_upsampler (nn.Module): The upsampler for the background. Default: None.
"""
def __init__(self, device,model_path, upscale=2, arch='clean', channel_multiplier=2, bg_upsampler=None):
self.upscale = upscale
self.bg_upsampler = bg_upsampler
# initialize model
self.device = device#torch.device('cuda' if torch.cuda.is_available() else 'cpu') if device is None else device
# initialize the GFP-GAN
if arch == 'clean':
self.gfpgan = GFPGANv1Clean(
out_size=512,
num_style_feat=512,
channel_multiplier=channel_multiplier,
decoder_load_path=None,
fix_decoder=False,
num_mlp=8,
input_is_latent=True,
different_w=True,
narrow=1,
sft_half=True)
self.face_helper = FaceRestoreHelper(
upscale,
face_size=512,
crop_ratio=(1, 1),
det_model='retinaface_resnet50',
save_ext='png',
use_parse=True,
device=self.device)
loadnet = torch.load(model_path, map_location=device)
#loadnet = torch.load(model_path)
if 'params_ema' in loadnet:
keyname = 'params_ema'
else:
keyname = 'params'
self.gfpgan.load_state_dict(loadnet[keyname], strict=True)
self.gfpgan.eval()
self.gfpgan = self.gfpgan.to(self.device)
print('GFPGAN model loaded')
@torch.no_grad()
def enhance_allimg(self, img, has_aligned=False, only_center_face=False, paste_back=True):
self.face_helper.clean_all()
import time
start = time.time()
if has_aligned: # the inputs are already aligned
img = cv2.resize(img, (512, 512))
self.face_helper.cropped_faces = [img]
else:
self.face_helper.read_image(img)
# get face landmarks for each face
self.face_helper.get_face_landmarks_5(only_center_face=only_center_face, eye_dist_threshold=5)
self.face_helper.align_warp_face()
end = time.time()
# face restoration
start = time.time()
for cropped_face in self.face_helper.cropped_faces:
# prepare data
start = time.time()
cropped_face_t = img2tensor(cropped_face / 255., bgr2rgb=True, float32=True)
normalize(cropped_face_t, (0.5, 0.5, 0.5), (0.5, 0.5, 0.5), inplace=True)
cropped_face_t = cropped_face_t.unsqueeze(0).to(self.device)
end = time.time()
try:
output = self.gfpgan(cropped_face_t, return_rgb=False)[0] # 15ms #NCHW
restored_face = tensor2img(output.squeeze(0), rgb2bgr=True, min_max=(-1, 1)) # 18msms
except RuntimeError as error:
print(f'\tFailed inference for GFPGAN: {error}.')
restored_face = cropped_face
start = time.time()
restored_face = restored_face.astype('uint8')
self.face_helper.add_restored_face(restored_face)
end = time.time()
end = time.time()
start = time.time()
if not has_aligned and paste_back:
# upsample the background
if self.bg_upsampler is not None:
# Now only support RealESRGAN for upsampling background
bg_img = self.bg_upsampler.enhance(img, outscale=self.upscale)[0]
else:
bg_img = None
self.face_helper.get_inverse_affine(None)
# paste each restored face to the input image
restored_img = self.face_helper.paste_faces_to_input_image(upsample_img=bg_img)
return self.face_helper.cropped_faces, self.face_helper.restored_faces, restored_img
else:
return self.face_helper.cropped_faces, self.face_helper.restored_faces, None
end = time.time()
@torch.no_grad()
def enhance(self, img, has_aligned=False, only_center_face=False, paste_back=True):
self.face_helper.clean_all()
if has_aligned: # the inputs are already aligned
# img = torch_resize(img)
img = cv2.resize(img, (512, 512))
self.face_helper.cropped_faces = [img]
else:
self.face_helper.read_image(img)
# get face landmarks for each face
self.face_helper.get_face_landmarks_5(only_center_face=only_center_face, eye_dist_threshold=5)
self.face_helper.align_warp_face()
start = time.time()
for cropped_face in self.face_helper.cropped_faces:
# prepare data
start = time.time()
cropped_face_t = img2tensor(cropped_face / 255., bgr2rgb=True, float32=True)
normalize(cropped_face_t, (0.5, 0.5, 0.5), (0.5, 0.5, 0.5), inplace=True)
cropped_face_t = cropped_face_t.unsqueeze(0).to(self.device)#([1, 3, 512, 512])
end = time.time()
try:
output = self.gfpgan(cropped_face_t, return_rgb=False)[0] #15ms #NCHW
restored_face = tensor2img(output.squeeze(0), rgb2bgr=True, min_max=(-1, 1)) #18msms
except RuntimeError as error:
print(f'\tFailed inference for GFPGAN: {error}.')
restored_face = cropped_face
start = time.time()
restored_face = restored_face.astype('uint8')
self.face_helper.add_restored_face(restored_face)
end = time.time()
if not has_aligned and paste_back:
# upsample the background
if self.bg_upsampler is not None:
# Now only support RealESRGAN for upsampling background
bg_img = self.bg_upsampler.enhance(img, outscale=self.upscale)[0]
else:
bg_img = None
self.face_helper.get_inverse_affine(None)
# paste each restored face to the input image
restored_img = self.face_helper.paste_faces_to_input_image(upsample_img=bg_img)
return self.face_helper.cropped_faces, self.face_helper.restored_faces, restored_img
else:
return self.face_helper.cropped_faces, self.face_helper.restored_faces, None
end = time.time()
print('paste_faces_to_input_image face: ', (end - start) * 1000)
def GFPGANInit(device,face_enhancement_path):
"""Inference demo for GFPGAN (for users).
"""
upscale = 1
# ------------------------ input & output ------------------------
import numpy as np
bg_upsampler = None
# ------------------------ set up GFPGAN restorer ------------------------
arch = 'clean'
channel_multiplier = 2
model_name = 'GFPGANv1.3'
model_path = face_enhancement_path
restorer = GFPGANer(
device = device,
model_path=model_path,
upscale=upscale,
arch=arch,
channel_multiplier=channel_multiplier,
bg_upsampler=bg_upsampler)
return restorer
def GFPGANInfer(img, restorer, aligned):
only_center_face = True
start = time.time()
if aligned:
cropped_faces, restored_faces, restored_img = restorer.enhance(
img, has_aligned=aligned, only_center_face=only_center_face, paste_back=True)
else:
cropped_faces, restored_faces, restored_img = restorer.enhance_allimg(
img, has_aligned=aligned, only_center_face=only_center_face, paste_back=True)
end = time.time()
if aligned==False:
return restored_img
else:
return restored_faces[0]