Spaces:
Running
on
A10G
Running
on
A10G
File size: 6,695 Bytes
320e465 |
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 |
import torch
import torch.nn as nn
import lpips
from model.vgg_arch import VGGFeatureExtractor
class PerceptualLoss(nn.Module):
"""Perceptual loss with commonly used style loss.
Args:
layer_weights (dict): The weight for each layer of vgg feature.
Here is an example: {'conv5_4': 1.}, which means the conv5_4
feature layer (before relu5_4) will be extracted with weight
1.0 in calculting losses.
vgg_type (str): The type of vgg network used as feature extractor.
Default: 'vgg19'.
use_input_norm (bool): If True, normalize the input image in vgg.
Default: True.
range_norm (bool): If True, norm images with range [-1, 1] to [0, 1].
Default: False.
perceptual_weight (float): If `perceptual_weight > 0`, the perceptual
loss will be calculated and the loss will multiplied by the
weight. Default: 1.0.
style_weight (float): If `style_weight > 0`, the style loss will be
calculated and the loss will multiplied by the weight.
Default: 0.
criterion (str): Criterion used for perceptual loss. Default: 'l1'.
"""
def __init__(self,
layer_weights,
vgg_type='vgg19',
use_input_norm=True,
range_norm=False,
perceptual_weight=1.0,
style_weight=0.,
criterion='l1'):
super(PerceptualLoss, self).__init__()
self.perceptual_weight = perceptual_weight
self.style_weight = style_weight
self.layer_weights = layer_weights
self.vgg = VGGFeatureExtractor(
layer_name_list=list(layer_weights.keys()),
vgg_type=vgg_type,
use_input_norm=use_input_norm,
range_norm=range_norm)
self.criterion_type = criterion
if self.criterion_type == 'l1':
self.criterion = torch.nn.L1Loss()
elif self.criterion_type == 'l2':
self.criterion = torch.nn.L2loss()
elif self.criterion_type == 'mse':
self.criterion = torch.nn.MSELoss(reduction='mean')
elif self.criterion_type == 'fro':
self.criterion = None
else:
raise NotImplementedError(f'{criterion} criterion has not been supported.')
def forward(self, x, gt):
"""Forward function.
Args:
x (Tensor): Input tensor with shape (n, c, h, w).
gt (Tensor): Ground-truth tensor with shape (n, c, h, w).
Returns:
Tensor: Forward results.
"""
# extract vgg features
x_features = self.vgg(x)
gt_features = self.vgg(gt.detach())
# calculate perceptual loss
if self.perceptual_weight > 0:
percep_loss = 0
for k in x_features.keys():
if self.criterion_type == 'fro':
percep_loss += torch.norm(x_features[k] - gt_features[k], p='fro') * self.layer_weights[k]
else:
percep_loss += self.criterion(x_features[k], gt_features[k]) * self.layer_weights[k]
percep_loss *= self.perceptual_weight
else:
percep_loss = None
# calculate style loss
if self.style_weight > 0:
style_loss = 0
for k in x_features.keys():
if self.criterion_type == 'fro':
style_loss += torch.norm(
self._gram_mat(x_features[k]) - self._gram_mat(gt_features[k]), p='fro') * self.layer_weights[k]
else:
style_loss += self.criterion(self._gram_mat(x_features[k]), self._gram_mat(
gt_features[k])) * self.layer_weights[k]
style_loss *= self.style_weight
else:
style_loss = None
return percep_loss, style_loss
def _gram_mat(self, x):
"""Calculate Gram matrix.
Args:
x (torch.Tensor): Tensor with shape of (n, c, h, w).
Returns:
torch.Tensor: Gram matrix.
"""
n, c, h, w = x.size()
features = x.view(n, c, w * h)
features_t = features.transpose(1, 2)
gram = features.bmm(features_t) / (c * h * w)
return gram
class LPIPSLoss(nn.Module):
def __init__(self,
loss_weight=1.0,
use_input_norm=True,
range_norm=False,):
super(LPIPSLoss, self).__init__()
self.perceptual = lpips.LPIPS(net="vgg", spatial=False).eval()
self.loss_weight = loss_weight
self.use_input_norm = use_input_norm
self.range_norm = range_norm
if self.use_input_norm:
# the mean is for image with range [0, 1]
self.register_buffer('mean', torch.Tensor([0.485, 0.456, 0.406]).view(1, 3, 1, 1))
# the std is for image with range [0, 1]
self.register_buffer('std', torch.Tensor([0.229, 0.224, 0.225]).view(1, 3, 1, 1))
def forward(self, pred, target):
if self.range_norm:
pred = (pred + 1) / 2
target = (target + 1) / 2
if self.use_input_norm:
pred = (pred - self.mean) / self.std
target = (target - self.mean) / self.std
lpips_loss = self.perceptual(target.contiguous(), pred.contiguous())
return self.loss_weight * lpips_loss.mean(), None
class AdversarialLoss(nn.Module):
r"""
Adversarial loss
https://arxiv.org/abs/1711.10337
"""
def __init__(self,
type='nsgan',
target_real_label=1.0,
target_fake_label=0.0):
r"""
type = nsgan | lsgan | hinge
"""
super(AdversarialLoss, self).__init__()
self.type = type
self.register_buffer('real_label', torch.tensor(target_real_label))
self.register_buffer('fake_label', torch.tensor(target_fake_label))
if type == 'nsgan':
self.criterion = nn.BCELoss()
elif type == 'lsgan':
self.criterion = nn.MSELoss()
elif type == 'hinge':
self.criterion = nn.ReLU()
def __call__(self, outputs, is_real, is_disc=None):
if self.type == 'hinge':
if is_disc:
if is_real:
outputs = -outputs
return self.criterion(1 + outputs).mean()
else:
return (-outputs).mean()
else:
labels = (self.real_label
if is_real else self.fake_label).expand_as(outputs)
loss = self.criterion(outputs, labels)
return loss
|