forked from dongdong4fei/DPE-MEF
-
Notifications
You must be signed in to change notification settings - Fork 0
/
models.py
149 lines (121 loc) · 4.28 KB
/
models.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
import torch
import torch.nn as nn
import torch.nn.functional as F
class DEM(nn.Module):
def __init__(self):
super(DEM, self).__init__()
self.inc_1 = DoubleConv(3, 16)
self.down1_1 = Down(16, 32)
self.down2_1 = Down(32, 64)
self.down3_1 = Down(64, 128)
self.inc_2 = DoubleConv(6, 32)
self.down1_2 = Down(32, 64)
self.down2_2 = Down(64, 128)
self.down3_2 = Down(128, 256)
factor = 2
# factor = 1
self.down4_1 = Down(128, 256 // factor)
self.down4_2 = Down(256, 512 // factor)
self.up1 = Up(256 * 4, 256 // factor)
self.up2 = Up(128 * 3, 128 // factor)
self.up3 = Up(64 * 3, 64 // factor)
self.up4 = Up(32 * 3, 32)
self.outc = OutConv(32, 1)
def forward(self, input):
img1 = input[:, 0:3, :, :]
img2 = input[:, 3:, :, :]
x1_1 = self.inc_1(img1)
x2_1 = self.down1_1(x1_1)
x3_1 = self.down2_1(x2_1)
x4_1 = self.down3_1(x3_1)
x5_1 = self.down4_1(x4_1)
x1_2 = self.inc_1(img2)
x2_2 = self.down1_1(x1_2)
x3_2 = self.down2_1(x2_2)
x4_2 = self.down3_1(x3_2)
x5_2 = self.down4_1(x4_2)
j1 = self.inc_2(input)
j2 = self.down1_2(j1)
j3 = self.down2_2(j2)
j4 = self.down3_2(j3)
j5 = self.down4_2(j4)
x = self.up1(torch.cat([x5_1, x5_2, j5], dim=1), torch.cat([x4_1, x4_2, j4], dim=1))
x = self.up2(x, torch.cat([x3_1, x3_2, j3], dim=1))
x = self.up3(x, torch.cat([x2_1, x2_2, j2], dim=1))
x = self.up4(x, torch.cat([x1_1, x1_2, j1], dim=1))
out = self.outc(x)
out = F.tanh(out)
out = out / 2 + 0.5
return out
class CEM(nn.Module):
def __init__(self):
super(CEM, self).__init__()
self.inc = DoubleConv(7, 32)
self.down1 = Down(32, 64)
self.down2 = Down(64, 128)
self.down3 = Down(128, 256)
factor = 2
# factor = 1
self.down4 = Down(256, 512 // factor)
self.up1 = Up(512, 256 // factor)
self.up2 = Up(256, 128 // factor)
self.up3 = Up(128, 64 // factor)
self.up4 = Up(64, 32)
self.outc = OutConv(32, 2)
def forward(self, x):
x1 = self.inc(x)
x2 = self.down1(x1)
x3 = self.down2(x2)
x4 = self.down3(x3)
x5 = self.down4(x4)
x = self.up1(x5, x4)
x = self.up2(x, x3)
x = self.up3(x, x2)
x = self.up4(x, x1)
out = self.outc(x)
out = F.tanh(out)
out = out / 2 + 0.5
return out
class DoubleConv(nn.Module):
def __init__(self, in_channels, out_channels, mid_channels=None):
super().__init__()
if not mid_channels:
mid_channels = out_channels
self.double_conv = nn.Sequential(
nn.Conv2d(in_channels, mid_channels, kernel_size=3, padding=1),
nn.BatchNorm2d(mid_channels),
nn.ReLU(inplace=True),
nn.Conv2d(mid_channels, out_channels, kernel_size=3, padding=1),
nn.BatchNorm2d(out_channels),
nn.ReLU(inplace=True)
)
def forward(self, x):
return self.double_conv(x)
class Down(nn.Module):
def __init__(self, in_channels, out_channels):
super().__init__()
self.maxpool_conv = nn.Sequential(
nn.MaxPool2d(2),
DoubleConv(in_channels, out_channels)
)
def forward(self, x):
return self.maxpool_conv(x)
class Up(nn.Module):
def __init__(self, in_channels, out_channels):
super().__init__()
self.up = nn.Upsample(scale_factor=2, mode='bilinear', align_corners=True)
self.conv = DoubleConv(in_channels, out_channels, in_channels // 2)
def forward(self, x1, x2):
x1 = self.up(x1)
diffY = x2.size()[2] - x1.size()[2]
diffX = x2.size()[3] - x1.size()[3]
x1 = F.pad(x1, [diffX // 2, diffX - diffX // 2,
diffY // 2, diffY - diffY // 2])
x = torch.cat([x2, x1], dim=1)
return self.conv(x)
class OutConv(nn.Module):
def __init__(self, in_channels, out_channels):
super(OutConv, self).__init__()
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=1)
def forward(self, x):
return self.conv(x)