-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmodel.py
212 lines (206 loc) · 8.02 KB
/
model.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
import torch
import torch.nn as nn
import numpy as np
class Bottleneck(nn.Module): #Basic模块
def __init__(self, inputs, outputs, stride=1, downsample=None):
super(Bottleneck, self).__init__()
self.conv1 = nn.Conv2d(inputs, outputs, kernel_size=1)
self.bn1 = nn.BatchNorm2d(outputs)
self.conv2 = nn.Conv2d(outputs, outputs, kernel_size=3, stride=stride, padding=1)
self.bn2 = nn.BatchNorm2d(outputs)
self.conv3 = nn.Conv2d(outputs, outputs * 4, kernel_size=1)
self.bn3 = nn.BatchNorm2d(outputs * 4)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
def forward(self, x):
residual = x
out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)
out = self.conv2(out)
out = self.bn2(out)
out = self.relu(out)
out = self.conv3(out)
out = self.bn3(out)
out = self.relu(out)
if self.downsample is not None:
residual = self.downsample(residual)
out += residual
out = self.relu(x)
return x
class BasicBlock(nn.Module): #Basic模块
expansion = 1
def __init__(self, inputs, outputs, stride=1):
super(BasicBlock, self).__init__()
self.conv1 = nn.Conv2d(inputs, outputs, kernel_size=3, stride=stride, padding=1)
self.bn1 = nn.BatchNorm2d(outputs)
self.relu = nn.ReLU(inplace=True)
self.conv2 = nn.Conv2d(outputs, outputs, kernel_size=3, stride=stride, padding=1)
self.bn2 = nn.BatchNorm2d(outputs)
self.downsample = downsample
self.stride = stride
def forward(self, x):
out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)
out = self.conv2(out)
out = self.bn2(out)
out = self.relu(out)
return out
class StageModule(nn.Module): #构建Stage
def __init__(self, inputs, outputs, first_channel):
super(StageModule, self).__init__()
self.inputs = inputs
self.outputs = outputs
self.branches = nn.ModuleList()
for i in range(self.inputs): # 首先通过四个BasicBlock模块
w = first_channel * (2 ** i) # 对应分支的通道数，2的倍数增加
branch = nn.Sequential(
BasicBlock(w, w),
BasicBlock(w, w),
BasicBlock(w, w),
BasicBlock(w, w)
)
self.branches.append(branch)
self.fuse_layers = nn.ModuleList()
for i in range(self.outputs):
self.fuse_layers.append(nn.ModuleList())
for j in range(self.inputs):
if i == j:
self.fuse_layers[-1].append(nn.Identity())
elif i < j: #输出分支小于输入分支时，需要对输入分支进行上采样
self.fuse_layers[-1].append(
nn.Sequential(
nn.Conv2d(c * (2 ** j), c * (2 ** i), kernel_size=1, stride=1),
nn.BatchNorm2d(c * (2 ** i)),
nn.Upsample(scale_factor=2.0 ** (j - i), mode='nearest')
)
)
else: #输出分支小于输入分支时，需要对输入分支进行下采样
ops = []
for k in range(i - j - 1):
ops.append(
nn.Sequential(
nn.Conv2d(c * (2 ** j), c * (2 ** j), kernel_size=3, stride=2, padding=1),
nn.BatchNorm2d(c * (2 ** j)),
nn.ReLU(inplace=True)
)
)
ops.append(
nn.Sequential(
nn.Conv2d(c * (2 ** j), c * (2 ** i), kernel_size=3, stride=2, padding=1),
nn.BatchNorm2d(c * (2 ** i))
)
)
self.fuse_layers[-1].append(nn.Sequential(*ops))
self.relu = nn.ReLU(inplace=True)
def forward(self, x):
# 每个分支通过对应的block
x = [branch(xi) for branch, xi in zip(self.branches, x)]
output = []
for i in range(len(self.fuse_layers)):
output.append(
self.relu(
sum([self.fuse_layers[i][j](x[j]) for j in range(len(self.branches))])
)
)
return output
class HRnet(nn.Module):
def __init__(self, first_channel = 32, num_joints = 27):
super(HRnet, self).__init__()
self.conv1 = nn.Conv2d(3, 64, kernel_size=3, stride=2, padding=1)
self.bn1 = nn.BatchNorm2d(64)
self.conv2 = nn.Conv2d(64, 64, kernel_size=3, stride=2, padding=1)
self.bn2 = nn.BatchNorm2d(64)
self.relu = nn.ReLU(inplace=True)
# 第一阶段
downsample = nn.Sequential( #通道需要变成四倍
nn.Conv2d(64, 256, kernel_size=1, stride=1, bias=False),
nn.BatchNorm2d(256)
)
self.layer1 = nn.Sequential(
Bottleneck(64, 64, downsample=downsample),
Bottleneck(256, 64),
Bottleneck(256, 64),
Bottleneck(256, 64)
)
self.transition1 = nn.ModuleList([
nn.Sequential(
nn.Conv2d(256, first_channel, kernel_size=3, stride=1, padding=1, bias=False),
nn.BatchNorm2d(first_channel, momentum=BN_MOMENTUM),
nn.ReLU(inplace=True)
),
nn.Sequential(
nn.Sequential( # 这里又使用一次Sequential是为了适配原项目中提供的权重
nn.Conv2d(256, first_channel * 2, kernel_size=3, stride=2, padding=1, bias=False),
nn.BatchNorm2d(first_channel * 2, momentum=BN_MOMENTUM),
nn.ReLU(inplace=True)
)
)
])
# 第二阶段
self.stage2 = nn.Sequential(
StageModule(2, 2, first_channel)
)
self.transition2 = nn.ModuleList([
nn.Identity(),
nn.Identity(),
nn.Sequential(
nn.Sequential(
nn.Conv2d(first_channel * 2, first_channel * 4, kernel_size=3, stride=2, padding=1),
nn.BatchNorm2d(first_channel * 4),
nn.ReLU(inplace=True)
)
)
])
# 第三阶段
self.stage3 = nn.Sequential(
StageModule(3, 3, first_channel),
StageModule(3, 3, first_channel),
StageModule(3, 3, first_channel),
StageModule(3, 3, first_channel)
)
self.transition3 = nn.ModuleList([
nn.Identity(),
nn.Identity(),
nn.Identity(),
nn.Sequential(
nn.Sequential(
nn.Conv2d(first_channel * 4, first_channel * 8, kernel_size=3, stride=2, padding=1),
nn.BatchNorm2d(first_channel * 8),
nn.ReLU(inplace=True)
)
)
])
# 第四阶段
self.stage4 = nn.Sequential(
StageModule(4, 4, first_channel),
StageModule(4, 4, first_channel),
StageModule(4, 1, first_channel)
)
self.out = nn.Conv2d(first_channel, num_joints, kernel_size=1, stride=1)
def forward(self, x):
x = self.conv1(x)
x = self.bn1(x)
x = self.relu(x)
x = self.conv2(x)
x = self.bn2(x)
x = self.relu(x)
x = self.layer1(x)
x = [trans(x) for trans in self.transition1]
x = self.stage2(x)
x = [
self.transition2[0](x[0]),
self.transition2[1](x[1]),
self.transition2[2](x[-1])
]
x = self.stage3(x)
x = [
self.transition3[0](x[0]),
self.transition3[1](x[1]),
self.transition3[2](x[2]),
self.transition3[3](x[-1]),
]
x = self.stage4(x)
x = self.out(x[0])
return x