non_local.py
6.09 KB
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
# Non-local block using embedded gaussian
# Code from
# https://github.com/AlexHex7/Non-local_pytorch/blob/master/Non-Local_pytorch_0.3.1/lib/non_local_embedded_gaussian.py
import torch
from torch import nn
from torch.nn import functional as F
class _NonLocalBlockND(nn.Module):
def __init__(self, in_channels, inter_channels=None, dimension=3, sub_sample=True, bn_layer=True):
super(_NonLocalBlockND, self).__init__()
assert dimension in [1, 2, 3]
self.dimension = dimension
self.sub_sample = sub_sample
self.in_channels = in_channels
self.inter_channels = inter_channels
if self.inter_channels is None:
self.inter_channels = in_channels // 2
if self.inter_channels == 0:
self.inter_channels = 1
if dimension == 3:
conv_nd = nn.Conv3d
max_pool_layer = nn.MaxPool3d(kernel_size=(1, 2, 2))
bn = nn.BatchNorm3d
elif dimension == 2:
conv_nd = nn.Conv2d
max_pool_layer = nn.MaxPool2d(kernel_size=(2, 2))
bn = nn.BatchNorm2d
else:
conv_nd = nn.Conv1d
max_pool_layer = nn.MaxPool1d(kernel_size=(2))
bn = nn.BatchNorm1d
self.g = conv_nd(in_channels=self.in_channels, out_channels=self.inter_channels,
kernel_size=1, stride=1, padding=0)
if bn_layer:
self.W = nn.Sequential(
conv_nd(in_channels=self.inter_channels, out_channels=self.in_channels,
kernel_size=1, stride=1, padding=0),
bn(self.in_channels)
)
nn.init.constant_(self.W[1].weight, 0)
nn.init.constant_(self.W[1].bias, 0)
else:
self.W = conv_nd(in_channels=self.inter_channels, out_channels=self.in_channels,
kernel_size=1, stride=1, padding=0)
nn.init.constant_(self.W.weight, 0)
nn.init.constant_(self.W.bias, 0)
self.theta = conv_nd(in_channels=self.in_channels, out_channels=self.inter_channels,
kernel_size=1, stride=1, padding=0)
self.phi = conv_nd(in_channels=self.in_channels, out_channels=self.inter_channels,
kernel_size=1, stride=1, padding=0)
if sub_sample:
self.g = nn.Sequential(self.g, max_pool_layer)
self.phi = nn.Sequential(self.phi, max_pool_layer)
def forward(self, x):
'''
:param x: (b, c, t, h, w)
:return:
'''
batch_size = x.size(0)
g_x = self.g(x).view(batch_size, self.inter_channels, -1)
g_x = g_x.permute(0, 2, 1)
theta_x = self.theta(x).view(batch_size, self.inter_channels, -1)
theta_x = theta_x.permute(0, 2, 1)
phi_x = self.phi(x).view(batch_size, self.inter_channels, -1)
f = torch.matmul(theta_x, phi_x)
f_div_C = F.softmax(f, dim=-1)
y = torch.matmul(f_div_C, g_x)
y = y.permute(0, 2, 1).contiguous()
y = y.view(batch_size, self.inter_channels, *x.size()[2:])
W_y = self.W(y)
z = W_y + x
return z
class NONLocalBlock1D(_NonLocalBlockND):
def __init__(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True):
super(NONLocalBlock1D, self).__init__(in_channels,
inter_channels=inter_channels,
dimension=1, sub_sample=sub_sample,
bn_layer=bn_layer)
class NONLocalBlock2D(_NonLocalBlockND):
def __init__(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True):
super(NONLocalBlock2D, self).__init__(in_channels,
inter_channels=inter_channels,
dimension=2, sub_sample=sub_sample,
bn_layer=bn_layer)
class NONLocalBlock3D(_NonLocalBlockND):
def __init__(self, in_channels, inter_channels=None, sub_sample=True, bn_layer=True):
super(NONLocalBlock3D, self).__init__(in_channels,
inter_channels=inter_channels,
dimension=3, sub_sample=sub_sample,
bn_layer=bn_layer)
class NL3DWrapper(nn.Module):
def __init__(self, block, n_segment):
super(NL3DWrapper, self).__init__()
self.block = block
self.nl = NONLocalBlock3D(block.bn3.num_features)
self.n_segment = n_segment
def forward(self, x):
x = self.block(x)
nt, c, h, w = x.size()
x = x.view(nt // self.n_segment, self.n_segment, c, h, w).transpose(1, 2) # n, c, t, h, w
x = self.nl(x)
x = x.transpose(1, 2).contiguous().view(nt, c, h, w)
return x
def make_non_local(net, n_segment):
import torchvision
import archs
if isinstance(net, torchvision.models.ResNet):
net.layer2 = nn.Sequential(
NL3DWrapper(net.layer2[0], n_segment),
net.layer2[1],
NL3DWrapper(net.layer2[2], n_segment),
net.layer2[3],
)
net.layer3 = nn.Sequential(
NL3DWrapper(net.layer3[0], n_segment),
net.layer3[1],
NL3DWrapper(net.layer3[2], n_segment),
net.layer3[3],
NL3DWrapper(net.layer3[4], n_segment),
net.layer3[5],
)
else:
raise NotImplementedError
if __name__ == '__main__':
from torch.autograd import Variable
import torch
sub_sample = True
bn_layer = True
img = Variable(torch.zeros(2, 3, 20))
net = NONLocalBlock1D(3, sub_sample=sub_sample, bn_layer=bn_layer)
out = net(img)
print(out.size())
img = Variable(torch.zeros(2, 3, 20, 20))
net = NONLocalBlock2D(3, sub_sample=sub_sample, bn_layer=bn_layer)
out = net(img)
print(out.size())
img = Variable(torch.randn(2, 3, 10, 20, 20))
net = NONLocalBlock3D(3, sub_sample=sub_sample, bn_layer=bn_layer)
out = net(img)
print(out.size())