注意力机制总结

最新推荐文章于 2024-06-02 11:45:10 发布

番茄就要炒鸡蛋

最新推荐文章于 2024-06-02 11:45:10 发布

阅读量1.4k

点赞数 2

分类专栏：注意力机制文章标签：深度学习

本文链接：https://blog.csdn.net/hejunran/article/details/119716155

版权

注意力机制专栏收录该内容

1 篇文章 0 订阅

订阅专栏

SE（Squeeze-and-Excitation）

BAM

CBAM

SimAM: A Simple, Parameter-Free Attention Module for Convolutional Neural Networks

总结

一：SE（Squeeze-and-Excitation）

1.1 SENet网络的创新点：

在于关注channel之间的关系，希望模型可以自动学习到不同channel特征的重要程度。

1.2 SE结构

如上图所示，为SE的结构，一个SEblock的过程分为 Squeeze(压缩) 和 Excitation(激发) 两个步骤：
1.Squeeze(压缩) 通过在Feature Map层上执行Global Average Pooling，得到当前Feature Map的全局压缩特征量；
2.Excitation(激发) 通过两层全连接的bottleneck结构得到Feature Map中每个通道的权值，并将加权后的Feature Map作为下一层网络的输入。

1.3 SE的操作细节

首先设输入的是一个 $H*W*C$ 的特征图，对它进行一个全局平均池化（核的大小 $H*W$ ），得到大小为 $C*1*1$ 的特征图。

$x2=globalavg(x1)$

然后经过两个全连接层进行非线性的处理。

第一个全连接层，FC1(C,C/16)

$x2=FC1(x2)$

第二个全连接层，FC2(C/16,C)

$x2=FC2(x2)$

两个全连接操作，先是降维，然后是升维，这样可以拟合通道之间的复杂相关性。之后再接上一个sigmoid层，得到一个 $C*1*1$ 的特征图，最后和原始的 $H*W*C$ 的特征图相乘。

$output=sigmoid(x2)*x1$

1.4 SE的应用

具体实现的代码：

resnet_se的实现

import torch.nn as nn
import math
import torch.utils.model_zoo as model_zoo


__all__ = ['SENet', 'se_resnet_18', 'se_resnet_34', 'se_resnet_50', 'se_resnet_101',
           'se_resnet_152']

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 BasicBlock(nn.Module):
    expansion = 1

    def __init__(self, inplanes, planes, stride=1, downsample=None):
        super(BasicBlock, self).__init__()
        self.conv1 = conv3x3(inplanes, planes, stride)
        self.bn1 = nn.BatchNorm2d(planes)
        self.relu = nn.ReLU(inplace=True)
        self.conv2 = conv3x3(planes, planes)
        self.bn2 = nn.BatchNorm2d(planes)
        self.downsample = downsample
        self.stride = stride

        if planes == 64:
            self.globalAvgPool = nn.AvgPool2d(56, stride=1)
        elif planes == 128:
            self.globalAvgPool = nn.AvgPool2d(28, stride=1)
        elif planes == 256:
            self.globalAvgPool = nn.AvgPool2d(14, stride=1)
        elif planes == 512:
            self.globalAvgPool = nn.AvgPool2d(7, stride=1)
        self.fc1 = nn.Linear(in_features=planes, out_features=round(planes / 16))
        self.fc2 = nn.Linear(in_features=round(planes / 16), out_features=planes)
        self.sigmoid = nn.Sigmoid()

    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.downsample is not None:
            residual = self.downsample(x)

        original_out = out
        out = self.globalAvgPool(out)
        out = out.view(out.size(0), -1)
        out = self.fc1(out)
        out = self.relu(out)
        out = self.fc2(out)
        out = self.sigmoid(out)
        out = out.view(out.size(0), out.size(1), 1, 1)
        out = out * original_out

        out += residual
        out = self.relu(out)

        return out


class Bottleneck(nn.Module):
    expansion = 4

    def __init__(self, inplanes, planes, stride=1, downsample=None):
        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,
                               padding=1, 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)
        if planes == 64:
            self.globalAvgPool = nn.AvgPool2d(56, stride=1)
        elif planes == 128:
            self.globalAvgPool = nn.AvgPool2d(28, stride=1)
        elif planes == 256:
            self.globalAvgPool = nn.AvgPool2d(14, stride=1)
        elif planes == 512:
            self.globalAvgPool = nn.AvgPool2d(7, stride=1)
        self.fc1 = nn.Linear(in_features=planes * 4, out_features=round(planes / 4))
        self.fc2 = nn.Linear(in_features=round(planes / 4), out_features=planes * 4)
        self.sigmoid = nn.Sigmoid()
        self.downsample = downsample
        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)
        out = self.relu(out)

        out = self.conv3(out)
        out = self.bn3(out)

        if self.downsample is not None:
            residual = self.downsample(x)

        original_out = out
        out = self.globalAvgPool(out)
        out = out.view(out.size(0), -1)
        out = self.fc1(out)
        out = self.relu(out)
        out = self.fc2(out)
        out = self.sigmoid(out)
        out = out.view(out.size(0),out.size(1),1,1)
        out = out * original_out

        out += residual
        out = self.relu(out)

        return out


class SENet(nn.Module):

    def __init__(self, block, layers, num_classes=1000):
        self.inplanes = 64
        super(SENet, self).__init__()
        self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3,
                               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)
        self.layer4 = self._make_layer(block, 512, layers[3], stride=2)
        self.avgpool = nn.AvgPool2d(7, stride=1)
        self.fc = nn.Linear(512 * block.expansion, num_classes)

        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_()

    def _make_layer(self, block, planes, blocks, stride=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))

        return nn.Sequential(*layers)

    def forward(self, x):
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.relu(x)
        x = self.maxpool(x)

        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        x = self.layer4(x)

        x = self.avgpool(x)
        x = x.view(x.size(0), -1)
        x = self.fc(x)

        return x


def se_resnet_18(pretrained=False, **kwargs):
    """Constructs a ResNet-18 model.
    Args:
        pretrained (bool): If True, returns a model pre-trained on ImageNet
    """
    model = SENet(BasicBlock, [2, 2, 2, 2], **kwargs)
    return model


def se_resnet_34(pretrained=False, **kwargs):
    """Constructs a ResNet-34 model.
    Args:
        pretrained (bool): If True, returns a model pre-trained on ImageNet
    """
    model = SENet(BasicBlock, [3, 4, 6, 3], **kwargs)
    return model


def se_resnet_50(pretrained=False, **kwargs):
    """Constructs a ResNet-50 model.
    Args:
        pretrained (bool): If True, returns a model pre-trained on ImageNet
    """
    model = SENet(Bottleneck, [3, 4, 6, 3], **kwargs)
    return model


def se_resnet_101(pretrained=False, **kwargs):
    """Constructs a ResNet-101 model.
    Args:
        pretrained (bool): If True, returns a model pre-trained on ImageNet
    """
    model = SENet(Bottleneck, [3, 4, 23, 3], **kwargs)
    return model


def se_resnet_152(pretrained=False, **kwargs):
    """Constructs a ResNet-152 model.
    Args:
        pretrained (bool): If True, returns a model pre-trained on ImageNet
    """
    model = SENet(Bottleneck, [3, 8, 36, 3], **kwargs)
    return model

if __name__ == '__main__':
    resnet_model=se_resnet_50(pretrained=False)
    print(resnet_model)

二：BAM （Bottlenet attention Module）

2.1BAM结构的创新点

它的优势：

第一有效的得到全局上下文信息，CNN堆积许多卷积层和池化层来获取上下文信息，虽然有效，但是增加了时间和空间复杂度。

第二忽略了底层特征，而BAM可以放在模块的开始，那么底层特征也能得到上下文信息

BAM结构通过两个分离的路径 channel和spatial，可以结合到任何前向传播卷积神经网络中，得到一个注意力图（Attention Map）。

2.2BAM结构

如上图所示，BAM由两个部分组成：channel和spatial分支组成。

channel attention branch

其中通道分支，由一个全局平均池化，两个全连接构成

spatial attention branch

其中空间分支，由两个1*1的卷积，两个3*3的空洞卷积

2.3BAM的实现细节

channel attention branch

首先设输入的是一个 $H*W*C$ 的特征图，对它进行一个全局平均池化（核的大小 $H*W$ ），得到大小为 $C*1*1$ 的特征图。

$x2=globalavg(x1)$

然后经过两个全连接层进行非线性的处理。

第一个全连接层，FC1(C,C/16)

$x2=FC1(x2)$

第二个全连接层，FC2(C/16,C)

$x2=FC2(x2)$

把得到的 $x2$ 特征图和空间分支得到特征图进行融合

spatial attention branch

首先设输入的是一个 $H*W*C$ 的特征图，

然后经过四次卷积操作：

第一次卷积：conv1（C，C/16，k=1，s=1）

$x1=relu(bn(conv1(x)))$

第二次卷积：conv2（C/16，C/16，k=3，p=4，s=4）

$x2=relu(bn(conv2(x1)))$

第三次卷积：conv3（C/16，C/16，k=3，p=4，s=4）

$x3=relu(bn(conv3(x2)))$

第四次卷积：conv4（C/16，C，k=1，s=1）

$x4=conv4(x3)$

最后两个通道进行融合：

$output1=sigmoid(channelx2*spatialx4)+1$

最后和原始特征图进行融合：

$output = x*output1$

这样就用BAM得到了一个空间的注意力特征图。

2.4BAM的应用

BAM可以嵌套resnet中。

代码实现：

import torch
import math
import torch.nn as nn
import torch.nn.functional as F

class Flatten(nn.Module):
    def forward(self, x):
        return x.view(x.size(0), -1)
class ChannelGate(nn.Module):
    def __init__(self, gate_channel, reduction_ratio=16, num_layers=1):
        super(ChannelGate, self).__init__()
        # self.gate_activation = gate_activation
        self.gate_c = nn.Sequential()
        self.gate_c.add_module( 'flatten', Flatten() )
        gate_channels = [gate_channel]
        gate_channels += [gate_channel // reduction_ratio] * num_layers
        gate_channels += [gate_channel]
        for i in range( len(gate_channels) - 2 ):
            self.gate_c.add_module( 'gate_c_fc_%d'%i, nn.Linear(gate_channels[i], gate_channels[i+1]) )
            self.gate_c.add_module( 'gate_c_bn_%d'%(i+1), nn.BatchNorm1d(gate_channels[i+1]) )
            self.gate_c.add_module( 'gate_c_relu_%d'%(i+1), nn.ReLU() )
        self.gate_c.add_module( 'gate_c_fc_final', nn.Linear(gate_channels[-2], gate_channels[-1]) )
    def forward(self, in_tensor):
        avg_pool = F.avg_pool2d( in_tensor, in_tensor.size(2), stride=in_tensor.size(2) )
        return self.gate_c( avg_pool ).unsqueeze(2).unsqueeze(3).expand_as(in_tensor)

class SpatialGate(nn.Module):
    def __init__(self, gate_channel, reduction_ratio=16, dilation_conv_num=2, dilation_val=4):
        super(SpatialGate, self).__init__()
        self.gate_s = nn.Sequential()
        self.gate_s.add_module( 'gate_s_conv_reduce0', nn.Conv2d(gate_channel, gate_channel//reduction_ratio, kernel_size=1))
        self.gate_s.add_module( 'gate_s_bn_reduce0',	nn.BatchNorm2d(gate_channel//reduction_ratio) )
        self.gate_s.add_module( 'gate_s_relu_reduce0',nn.ReLU() )
        for i in range( dilation_conv_num ):
            self.gate_s.add_module( 'gate_s_conv_di_%d'%i, nn.Conv2d(gate_channel//reduction_ratio, gate_channel//reduction_ratio, kernel_size=3, \
						padding=dilation_val, dilation=dilation_val) )
            self.gate_s.add_module( 'gate_s_bn_di_%d'%i, nn.BatchNorm2d(gate_channel//reduction_ratio) )
            self.gate_s.add_module( 'gate_s_relu_di_%d'%i, nn.ReLU() )
        self.gate_s.add_module( 'gate_s_conv_final', nn.Conv2d(gate_channel//reduction_ratio, 1, kernel_size=1) )
    def forward(self, in_tensor):
        return self.gate_s( in_tensor ).expand_as(in_tensor)
class BAM(nn.Module):
    def __init__(self, gate_channel):
        super(BAM, self).__init__()
        self.channel_att = ChannelGate(gate_channel)
        self.spatial_att = SpatialGate(gate_channel)
    def forward(self,in_tensor):
        att = 1 + F.sigmoid( self.channel_att(in_tensor) * self.spatial_att(in_tensor) )
        return att * in_tensor

三：CBAM

3.1CBAM结构的创新点

作者将注意力过程分为两个独立的部分，通道注意力模块和空间注意力模块。这样不仅可以节约参数和计算力，而且保证了其可以作为即插即用的模块集成到现有的网络架构中去。

3.2CBAM结构

CBAM由两个部分组成

3.3CBAM的实现细节

CBAM

channel attention branch

首先设输入的是一个 $H*W*C$ 的特征图，对它进行一个全局平均池化（核的大小 $H*W$ ）和一个全局最大值池化，得到两个大小为 $C*1*1$ 的特征图。

$x11=globalavgpool(x)$

$x12=globalmaxpool(x)$

然后经过两个全连接层进行非线性的处理。

第一个全连接层，FC1(C,C/16)

$x21=FC1(x11)$

$x22=FC1(x12)$

第二个全连接层，FC2(C/16,C)

$x31=FC2(x21)$

$x32=FC2(x22)$

把得到的两个特征图进行融合后，经过非线性变换后再与原始特征图进行融合：

$output1=sigmoid(x31+x32)*x$

spatial attention branch

其输入是通道分支的输出，首先进行两次通道维度的操作：

        avg_out = torch.mean(output1, dim=1, keepdim=True)
        max_out, _ = torch.max(output1, dim=1, keepdim=True)
        x1 = torch.cat([avg_out, max_out], dim=1)

然后进行一个卷积，卷积核的大小为7*7或者3*3

$x2=conv1(x1)$

经过非线性变换后再与原始特征图进行融合：

$output=sigmoid(x2)*output1$

3.4CBAM的应用

CBAM可以嵌套到一些主干网络中

代码实现：

import torch
import math
import torch.nn as nn
import torch.nn.functional as F

class BasicConv(nn.Module):
    def __init__(self, in_planes, out_planes, kernel_size, stride=1, padding=0, dilation=1, groups=1, relu=True, bn=True, bias=False):
        super(BasicConv, self).__init__()
        self.out_channels = out_planes
        self.conv = nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, padding=padding, dilation=dilation, groups=groups, bias=bias)
        self.bn = nn.BatchNorm2d(out_planes,eps=1e-5, momentum=0.01, affine=True) if bn else None
        self.relu = nn.ReLU() if relu else None

    def forward(self, x):
        x = self.conv(x)
        if self.bn is not None:
            x = self.bn(x)
        if self.relu is not None:
            x = self.relu(x)
        return x

class Flatten(nn.Module):
    def forward(self, x):
        return x.view(x.size(0), -1)

class ChannelGate(nn.Module):
    def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg', 'max']):
        super(ChannelGate, self).__init__()
        self.gate_channels = gate_channels
        self.mlp = nn.Sequential(
            Flatten(),
            nn.Linear(gate_channels, gate_channels // reduction_ratio),
            nn.ReLU(),
            nn.Linear(gate_channels // reduction_ratio, gate_channels)
            )
        self.pool_types = pool_types
    def forward(self, x):
        channel_att_sum = None
        for pool_type in self.pool_types:
            if pool_type=='avg':
                avg_pool = F.avg_pool2d( x, (x.size(2), x.size(3)), stride=(x.size(2), x.size(3)))
                channel_att_raw = self.mlp( avg_pool )
            elif pool_type=='max':
                max_pool = F.max_pool2d( x, (x.size(2), x.size(3)), stride=(x.size(2), x.size(3)))
                channel_att_raw = self.mlp( max_pool )
            elif pool_type=='lp':
                lp_pool = F.lp_pool2d( x, 2, (x.size(2), x.size(3)), stride=(x.size(2), x.size(3)))
                channel_att_raw = self.mlp( lp_pool )
            elif pool_type=='lse':
                # LSE pool only
                lse_pool = logsumexp_2d(x)
                channel_att_raw = self.mlp( lse_pool )

            if channel_att_sum is None:
                channel_att_sum = channel_att_raw
            else:
                channel_att_sum = channel_att_sum + channel_att_raw

        scale = F.sigmoid( channel_att_sum ).unsqueeze(2).unsqueeze(3).expand_as(x)
        return x * scale

def logsumexp_2d(tensor):
    tensor_flatten = tensor.view(tensor.size(0), tensor.size(1), -1)
    s, _ = torch.max(tensor_flatten, dim=2, keepdim=True)
    outputs = s + (tensor_flatten - s).exp().sum(dim=2, keepdim=True).log()
    return outputs

class ChannelPool(nn.Module):
    def forward(self, x):
        return torch.cat( (torch.max(x,1)[0].unsqueeze(1), torch.mean(x,1).unsqueeze(1)), dim=1 )

class SpatialGate(nn.Module):
    def __init__(self):
        super(SpatialGate, self).__init__()
        kernel_size = 7
        self.compress = ChannelPool()
        self.spatial = BasicConv(2, 1, kernel_size, stride=1, padding=(kernel_size-1) // 2, relu=False)
    def forward(self, x):
        x_compress = self.compress(x)
        x_out = self.spatial(x_compress)
        scale = F.sigmoid(x_out) # broadcasting
        return x * scale

class CBAM(nn.Module):
    def __init__(self, gate_channels, reduction_ratio=16, pool_types=['avg', 'max'], no_spatial=False):
        super(CBAM, self).__init__()
        self.ChannelGate = ChannelGate(gate_channels, reduction_ratio, pool_types)
        self.no_spatial=no_spatial
        if not no_spatial:
            self.SpatialGate = SpatialGate()
    def forward(self, x):
        x_out = self.ChannelGate(x)
        if not self.no_spatial:
            x_out = self.SpatialGate(x_out)
        return x_out