-
Notifications
You must be signed in to change notification settings - Fork 1
/
DCSA-Net.py
443 lines (343 loc) · 17 KB
/
DCSA-Net.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
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
import torch
from torch import nn
from torch.nn import functional as F
import math
import torch.utils.model_zoo as model_zoo
from torch.utils.checkpoint import checkpoint
from thop import profile
from torchstat import stat
from Dynamic_Convolution import Involution2d
from Dynamic_Selfattention import Second_feature_filtering_att
from CIAM import CIA
model_urls = {
'rednet50': 'file:///C:/Users/Administrator/Desktop/rednet50.pth',
}
class Invo_Plus_Net(nn.Module):
# Invo_filtering + context++(ASPP+CIA) + connection++
def __init__(self, num_class=37, pretrained=False):
super(Invo_Plus_Net, self).__init__()
layers = [3, 4, 6, 3]
block = Bottleneck
block_second = Bottleneck_att
transblock = TransBasicBlock
# RGB image branch
self.inplanes = 64
# self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3, bias=False)
self.invo1 = Involution2d(3, 64, kernel_size=7, stride=2, padding=3, reduce_ratio=4,
bias=False)
self.bn1 = nn.BatchNorm2d(64)
self.relu = nn.ReLU(inplace=True)
self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
self.layer1 = self._make_layer(block, 64, layers[0])
self.layer2 = self._make_layer(block, 128, layers[1], stride=2)
self.layer3 = self._make_layer(block, 256, layers[2], stride=2) # use PSPNet extractors
self.layer4 = self._make_layer(block, 512, layers[3], stride=2)
# depth image branch
self.inplanes = 64
#self.conv1_d = nn.Conv2d(1, 64, kernel_size=7, stride=2, padding=3, bias=False)
self.invo1_d = Involution2d(1, 64, kernel_size=7, stride=2, padding=3, reduce_ratio=4,
bias=False)
self.bn1_d = nn.BatchNorm2d(64)
self.relu_d = nn.ReLU(inplace=True)
self.maxpool_d = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
self.layer1_d = self._make_layer(block, 64, layers[0])
self.layer2_d = self._make_layer(block, 128, layers[1], stride=2)
self.layer3_d = self._make_layer(block, 256, layers[2], stride=2)
self.layer4_d = self._make_layer(block, 512, layers[3], stride=2)
# merge branch
self.maxpool_m = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
# downsamping
self.maxpool_down4x = nn.MaxPool2d(kernel_size=4, stride=4)
self.maxpool_down8x = nn.MaxPool2d(kernel_size=8, stride=8)
# adjust channel
self.conv_1024_512 = nn.Conv2d(1024,512, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_1024_256 = nn.Conv2d(1024,256, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_1024_64 = nn.Conv2d(1024,64, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_512_256 = nn.Conv2d(512,256, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_512_64 = nn.Conv2d(512,64, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_256_64 = nn.Conv2d(256,64, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_64_256 = nn.Conv2d(64,256, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_64_512 = nn.Conv2d(64,512, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_64_1024 = nn.Conv2d(64,1024, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_256_512 = nn.Conv2d(256,512, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_256_1024 = nn.Conv2d(256,1024, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_512_1024 = nn.Conv2d(512,1024, kernel_size=1, stride=1, padding=0, bias=False)
self.inplanes = 64
self.layer1_m = self._make_layer(block_second, 64, layers[0])
self.layer2_m = self._make_layer(block_second, 128, layers[1], stride=2)
self.layer3_m = self._make_layer(block_second, 256, layers[2], stride=2)
self.layer4_m = self._make_layer(block_second, 512, layers[3], stride=2)
# agant module
self.agant0 = self._make_agant_layer(64, 64)
self.agant1 = self._make_agant_layer(64*4, 64)
self.agant2 = self._make_agant_layer(128*4, 128)
self.agant3 = self._make_agant_layer(256*4, 256)
# self.agant4 = self._make_agant_layer(512*4, 512)
# ASPP + CIA
self.aspp = ASPP(256*4)
self.cia = CIA(256*4, 256, 256)
#transpose layer
self.inplanes = 512
self.deconv1 = self._make_transpose(transblock, 256, 6, stride=2)
self.deconv2 = self._make_transpose(transblock, 128, 4, stride=2)
self.deconv3 = self._make_transpose(transblock, 64, 3, stride=2)
self.deconv4 = self._make_transpose(transblock, 64, 3, stride=2)
# final blcok
self.inplanes = 64
self.final_conv = self._make_transpose(transblock, 64, 3)
self.final_deconv = nn.ConvTranspose2d(self.inplanes, num_class, kernel_size=2,
stride=2, padding=0, bias=True)
self.out5_conv = nn.Conv2d(256, num_class, kernel_size=1, stride=1, bias=True)
self.out4_conv = nn.Conv2d(128, num_class, kernel_size=1, stride=1, bias=True)
self.out3_conv = nn.Conv2d(64, num_class, kernel_size=1, stride=1, bias=True)
self.out2_conv = nn.Conv2d(64, num_class, kernel_size=1, stride=1, bias=True)
# weight initial
for m in self.modules():
if isinstance(m, nn.Conv2d):
n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels
m.weight.data.normal_(0, math.sqrt(2. / n))
elif isinstance(m, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
if pretrained:
self._load_resnet_pretrained()
def encoder(self, rgb, depth):
rgb = self.invo1(rgb) # 输入3 输出64 7*7卷积
rgb = self.bn1(rgb)
rgb = self.relu(rgb)
depth = self.invo1_d(depth) # 输入1 输出64
depth = self.bn1_d(depth)
depth = self.relu_d(depth)
m0 = rgb + depth # 128
rgb = self.maxpool(rgb)
depth = self.maxpool_d(depth)
m = self.maxpool_m(m0)
# block 1
rgb = self.layer1(rgb) # 只有3*3内卷
depth = self.layer1_d(depth) # 只有3*3内卷
m = self.layer1_m(m) # layer1_m是3*3内卷+自注意力
m1 = m + rgb + depth # 64
# block 2
rgb = self.layer2(rgb)
depth = self.layer2_d(depth)
m = self.layer2_m(m1)
m2 = m + rgb + depth # 32
# block 3
rgb = self.layer3(rgb) # 1 1024 16 16
depth = self.layer3_d(depth)
m = self.layer3_m(m2)
m3 = m + rgb + depth # 16
# connection++
m3_up_128 = self.conv_1024_64(F.interpolate(m3, size=m0.size()[2:], mode='bilinear', align_corners=True)) # 1 64 128 128
m3_up_64 = self.conv_1024_256(F.interpolate(m3, size=m1.size()[2:], mode='bilinear', align_corners=True)) # 1 256 64 64
m3_up_32 = self.conv_1024_512(F.interpolate(m3, size=m2.size()[2:], mode='bilinear', align_corners=True)) # 1 512 32 32
m2_up_128 = self.conv_512_64(F.interpolate(m2, size=m0.size()[2:], mode='bilinear', align_corners=True)) # 1 64 128 128
m2_up_64 = self.conv_512_256(F.interpolate(m2, size=m1.size()[2:], mode='bilinear', align_corners=True)) # 1 256 64 64
m1_up_128 = self.conv_256_64(F.interpolate(m1, size=m0.size()[2:], mode='bilinear', align_corners=True)) # 1 64 128 128
m0_down_64 = self.conv_64_256(self.maxpool_m(m0)) # 1 256 64 64
m0_down_32 = self.conv_64_512(self.maxpool_down4x(m0)) # 1 512 32 32
m0_down_16 = self.conv_64_1024(self.maxpool_down8x(m0)) # 1 1024 16 16
m1_down_32 = self.conv_256_512(self.maxpool_m(m1)) # 1 512 32 32
m1_down_16 = self.conv_256_1024(self.maxpool_down4x(m1)) # 1 1024 16 16
m2_down_16 = self.conv_512_1024(self.maxpool_m(m2)) # 1 1024 16 16
m0 = m0 + m3_up_128 + m2_up_128 + m1_up_128 # channel = 64 1 64 128 128
m1 = m1 + m0_down_64 + m3_up_64 + m2_up_64 # channel = 256 1 256 64 64
m2 = m2 + m0_down_32 + m3_up_32 + m1_down_32 # channel = 512 1 512 32 32
m3 = m0_down_16 + m1_down_16 + m2_down_16 # channel = 1024 1 1024 16 16
# ASPP + CIA
# m3_aspp = self.aspp(m3)
m3_cia = self.cia(m3) # 1 256 16 16
m3 = m3_cia
return m0, m1, m2, m3 # channel of m is 1024
def decoder(self, fuse0, fuse1, fuse2, fuse3):
# agent3 = self.agant3(fuse3)
# print(agent3.size()) [1, 256, 16, 16]
# 0:1 64 128 128 1:1 256 64 64 2:1 512 32 32 3:1 256 16 16
# upsample 2
x = self.deconv2(fuse3) # 1 128 32 32
x = x + self.agant2(fuse2) # 1 128 32 32
# print(x.size()) [1, 128, 32, 32]
# upsample 3
x = self.deconv3(x)
x = x + self.agant1(fuse1)
# print(x.size()) [1, 64, 64, 64]
# upsample 4
x = self.deconv4(x)
x = x + self.agant0(fuse0)
# print(x.size()) [1, 64, 128, 128]
# final
x = self.final_conv(x) # 1 64 128 128
out = self.final_deconv(x) # 1 6 256 256
return out
def forward(self, rgb, depth, phase_checkpoint=False):
fuses = self.encoder(rgb, depth) # 0:1 64 128 128 1:1 256 64 64 2:1 512 32 32 3:1 256 16 16
m = self.decoder(*fuses)
return m
def _make_layer(self, block, planes, blocks, stride=1, dilation=1):
downsample = None
if stride != 1 or self.inplanes != planes * block.expansion:
downsample = nn.Sequential(
nn.Conv2d(self.inplanes, planes * block.expansion,
kernel_size=1, stride=stride, bias=False),
nn.BatchNorm2d(planes * block.expansion),
)
layers = []
layers.append(block(self.inplanes, planes, stride, downsample))
self.inplanes = planes * block.expansion
for i in range(1, blocks):
layers.append(block(self.inplanes, planes, dilation=dilation))
return nn.Sequential(*layers)
def _make_agant_layer(self, inplanes, planes):
layers = nn.Sequential(
nn.Conv2d(inplanes, planes, kernel_size=1,
stride=1, padding=0, bias=False),
nn.BatchNorm2d(planes),
nn.ReLU(inplace=True)
)
return layers
def _make_transpose(self, block, planes, blocks, stride=1):
upsample = None
if stride != 1:
upsample = nn.Sequential(
nn.ConvTranspose2d(self.inplanes, planes,
kernel_size=2, stride=stride,
padding=0, bias=False),
nn.BatchNorm2d(planes),
)
elif self.inplanes != planes:
upsample = nn.Sequential(
nn.Conv2d(self.inplanes, planes,
kernel_size=1, stride=stride, bias=False),
nn.BatchNorm2d(planes),
)
layers = []
for i in range(1, blocks):
layers.append(block(self.inplanes, self.inplanes))
layers.append(block(self.inplanes, planes, stride, upsample))
self.inplanes = planes
return nn.Sequential(*layers)
def _load_resnet_pretrained(self):
pretrain_dict = model_zoo.load_url(model_urls['rednet50'])
model_dict = {}
state_dict = self.state_dict()
for k, v in pretrain_dict.items():
# print('%%%%% ', k)
if k in state_dict:
if k.startswith('invo1'):
model_dict[k] = v
# print('##### ', k)
model_dict[k.replace('invo1', 'invo1_d')] = torch.mean(v, 1).data. \
view_as(state_dict[k.replace('invo1', 'invo1_d')])
elif k.startswith('bn1'):
model_dict[k] = v
model_dict[k.replace('bn1', 'bn1_d')] = v
elif k.startswith('layer'):
model_dict[k] = v
model_dict[k[:6]+'_d'+k[6:]] = v
model_dict[k[:6]+'_m'+k[6:]] = v
state_dict.update(model_dict)
self.load_state_dict(state_dict)
class Bottleneck(nn.Module):
expansion = 4
def __init__(self, inplanes, planes, stride=1, downsample=None, dilation=1):
super(Bottleneck, self).__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
self.bn1 = nn.BatchNorm2d(planes)
# self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride, dilation=dilation, padding=dilation, bias=False)
self.invo2 = Involution2d(planes, planes, kernel_size=3, stride=stride, groups=16, reduce_ratio=4, dilation=dilation, padding=dilation, bias=False)
self.bn2 = nn.BatchNorm2d(planes)
self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
self.bn3 = nn.BatchNorm2d(planes * 4)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride
def forward(self, x):
residual = x
out = self.conv1(x) # 1*1卷积
out = self.bn1(out)
out = self.relu(out)
out = self.invo2(out) # 3*3内卷
out = self.bn2(out)
out = self.relu(out)
out = self.conv3(out)
out = self.bn3(out)
if self.downsample is not None:
residual = self.downsample(x)
out += residual # 残差
out = self.relu(out)
return out
class Bottleneck_att(nn.Module):
expansion = 4
def __init__(self, inplanes, planes, stride=1, downsample=None, dilation=1):
super(Bottleneck_att, self).__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
self.bn1 = nn.BatchNorm2d(planes)
# self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride, dilation=dilation, padding=dilation, bias=False)
self.invo2 = Involution2d(planes, planes, kernel_size=3, stride=stride, groups=16, reduce_ratio=4, dilation=dilation, padding=dilation, bias=False)
self.bn2 = nn.BatchNorm2d(planes)
self.second_filtering = Second_feature_filtering_att(planes)
self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
self.bn3 = nn.BatchNorm2d(planes * 4)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride
def forward(self, x):
residual = x
out = self.conv1(x) # 1*1卷积
out = self.bn1(out)
out = self.relu(out)
out = self.invo2(out) # 3*3内卷
out = self.bn2(out)
out = self.relu(out)
out = self.second_filtering(out) # 纯的通道和空间注意力
out = self.conv3(out)
out = self.bn3(out)
if self.downsample is not None:
residual = self.downsample(x)
out += residual # 残差
out = self.relu(out)
return out
def conv3x3(in_planes, out_planes, stride=1):
"3x3 convolution with padding"
return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride,
padding=1, bias=False)
class TransBasicBlock(nn.Module):
expansion = 1
def __init__(self, inplanes, planes, stride=1, upsample=None, **kwargs):
super(TransBasicBlock, self).__init__()
self.conv1 = conv3x3(inplanes, inplanes)
self.bn1 = nn.BatchNorm2d(inplanes)
self.relu = nn.ReLU(inplace=True)
if upsample is not None and stride != 1:
self.conv2 = nn.ConvTranspose2d(inplanes, planes,
kernel_size=3, stride=stride, padding=1,
output_padding=1, bias=False)
else:
self.conv2 = conv3x3(inplanes, planes, stride)
self.bn2 = nn.BatchNorm2d(planes)
self.upsample = upsample
self.stride = stride
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)
if self.upsample is not None:
residual = self.upsample(x)
out += residual
out = self.relu(out)
return out
if __name__ == "__main__":
model = Invo_Plus_Net(num_class=6, pretrained=False)
model.eval()
img = torch.randn(1,3,256,256)
dsm = torch.randn(1,1,256,256)
output = model(img,dsm)
print(output.size())
flops, params = profile(model, inputs=(img, dsm))
print("flops:",flops / 1000000000 , "G")
print("params",params / 1000000 , "M")
# flops: 21.109898144 G Invo_filtering + context++ + connection++
# params 27.118835 M