人脸检测 Retinaface - Backbone部分（MobileNet v1）
人脸检测 Retinaface - Backbone部分（MobileNet v1）
flyfish
MobileNetv1的原理部分
MobileNetv1的原始代码
class MobileNetv1(nn.Module):
def __init__(self):
super(MobileNetv1, self).__init__()
self.model = nn.Sequential(
conv_bn( 3, 32, 2),
conv_dw( 32, 64, 1),
conv_dw( 64, 128, 2),
conv_dw(128, 128, 1),
conv_dw(128, 256, 2),
conv_dw(256, 256, 1),
conv_dw(256, 512, 2),
conv_dw(512, 512, 1),
conv_dw(512, 512, 1),
conv_dw(512, 512, 1),
conv_dw(512, 512, 1),
conv_dw(512, 512, 1),
conv_dw(512, 1024, 2),
conv_dw(1024, 1024, 1),
nn.AvgPool2d(7),
)
self.fc = nn.Linear(1024, 1000)
def forward(self, x):
x = self.model(x)
x = x.view(-1, 1024)
x = self.fc(x)
return
RetinaFace 更改的MobileNetv1的代码
class MobileNetV1(nn.Module):
def __init__(self):
super(MobileNetV1, self).__init__()
self.stage1 = nn.Sequential(
conv_bn(3, 8, 2, leaky = 0.1), # 3
conv_dw(8, 16, 1), # 7
conv_dw(16, 32, 2), # 11
conv_dw(32, 32, 1), # 19
conv_dw(32, 64, 2), # 27
conv_dw(64, 64, 1), # 43
)
self.stage2 = nn.Sequential(
conv_dw(64, 128, 2), # 43 + 16 = 59
conv_dw(128, 128, 1), # 59 + 32 = 91
conv_dw(128, 128, 1), # 91 + 32 = 123
conv_dw(128, 128, 1), # 123 + 32 = 155
conv_dw(128, 128, 1), # 155 + 32 = 187
conv_dw(128, 128, 1), # 187 + 32 = 219
)
self.stage3 = nn.Sequential(
conv_dw(128, 256, 2), # 219 +3 2 = 241
conv_dw(256, 256, 1), # 241 + 64 = 301
)
self.avg = nn.AdaptiveAvgPool2d((1,1))
self.fc = nn.Linear(256, 1000)
def forward(self, x):
x = self.stage1(x)
x = self.stage2(x)
x = self.stage3(x)
x = self.avg(x)
# x = self.model(x)
x = x.view(-1, 256)
x = self.fc(x)
return x
结论：从32到1024 每个数都乘以0.25 也就是通道乘子是0.25
RetinaFace更改的MobileNet v1的网络架构
(body): IntermediateLayerGetter(
(stage1): Sequential(
(0): Sequential(
(0): Conv2d(3, 8, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(1): BatchNorm2d(8, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
)
(1): Sequential(
(0): Conv2d(8, 8, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=8, bias=False)
(1): BatchNorm2d(8, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(8, 16, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
(2): Sequential(
(0): Conv2d(16, 16, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=16, bias=False)
(1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(16, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
(3): Sequential(
(0): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)
(1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(32, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
(4): Sequential(
(0): Conv2d(32, 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=32, bias=False)
(1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(32, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
(5): Sequential(
(0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=64, bias=False)
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
)
(stage2): Sequential(
(0): Sequential(
(0): Conv2d(64, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=64, bias=False)
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(64, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
(1): Sequential(
(0): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=128, bias=False)
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
(2): Sequential(
(0): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=128, bias=False)
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
(3): Sequential(
(0): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=128, bias=False)
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
(4): Sequential(
(0): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=128, bias=False)
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
(5): Sequential(
(0): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=128, bias=False)
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
)
(stage3): Sequential(
(0): Sequential(
(0): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=128, bias=False)
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
(1): Sequential(
(0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False)
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
)
)
MobileNet v1原始的网络架构
MobileNet_原始(
(model): Sequential(
(0): Sequential(
(0): Conv2d(3, 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(1): Sequential(
(0): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)
(1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(32, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(2): Sequential(
(0): Conv2d(64, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=64, bias=False)
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(64, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(3): Sequential(
(0): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=128, bias=False)
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(4): Sequential(
(0): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=128, bias=False)
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(5): Sequential(
(0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False)
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(6): Sequential(
(0): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=256, bias=False)
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(7): Sequential(
(0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(8): Sequential(
(0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(9): Sequential(
(0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(10): Sequential(
(0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(11): Sequential(
(0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(12): Sequential(
(0): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=512, bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(512, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(13): Sequential(
(0): Conv2d(1024, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1024, bias=False)
(1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
(3): Conv2d(1024, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): ReLU(inplace=True)
)
(14): AvgPool2d(kernel_size=7, stride=7, padding=0)
)
(fc): Linear(in_features=1024, out_features=1000, bias=True)
)
Backbone是如何与FPN衔接的
我们改造的mobilenet v1 有三个stage
这三个stage都有输出，而FPN就是分别接在这三个stage之后
stage1与FPN的衔接
stage1的最后Sequential
(5): Sequential(
(0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=64, bias=False)
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
我们看输出结果是64,那与它对接的也是64,如下，正好stage1输出64，FPN的output1输入是64,
(output1): Sequential(
(0): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
)
stage2与FPN的衔接
stage2的最后Sequential是128
(5): Sequential(
(0): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=128, bias=False)
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
那与它对接的也是128,如下，正好stage2输出128，FPN的output2输入是128,
(output2): Sequential(
(0): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
)
stage3与FPN的衔接
stage3的最后Sequential是256
(1): Sequential(
(0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False)
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
(3): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(4): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(5): LeakyReLU(negative_slope=0.1, inplace=True)
)
FPN的对接
(output3): Sequential(
(0): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.1, inplace=True)
)