学习笔记1——常用的注意力机制（即插即用）_cam注意力机制

本文链接：https://blog.youkuaiyun.com/2401_86367018/article/details/142031987

    )

def forward(self, x):
    b, c, _, _ = x.size()
    y = self.avg_pool(x).view(b, c)
    y = self.fc(y).view(b, c, 1, 1)
    return x * y.expand_as(x)


2、CA模块（Coordinate attention）


论文：[https://arxiv.org/abs/2103.02907]( )


参考：[CA(Coordinate attention) 注意力机制 - 知乎 (zhihu.com)]( ) 注意力机制 - 知乎 (zhihu.com)")


[CVPR 2021 | 即插即用！ CA：新注意力机制，助力分类/检测/分割涨点！\_Amusi（CVer）的博客-优快云博客]( )


![](https://img-blog.csdnimg.cn/232b3dc1ed854fcfb97607d2f71ad7cd.png)


[CVPR 2021 | 即插即用！ CA：新注意力机制，助力分类/检测/分割涨点！\_Amusi（CVer）的博客-优快云博客]( )CA对宽度和高度两个方向分别全局平均池化，分别获得在宽度和高度两个方向的特征图，然后将两个方向的特征图concat，然后送入共享卷积将维度降为C/r，再通过批量归一化处理和激活函数后得到特征图。

import torch
from torch import nn

class CA_Block(nn.Module):
def init(self, channel, h, w, reduction=16):
super(CA_Block, self).init()

    self.h = h
    self.w = w

    self.avg_pool_x = nn.AdaptiveAvgPool2d((h, 1))
    self.avg_pool_y = nn.AdaptiveAvgPool2d((1, w))

    self.conv_1x1 = nn.Conv2d(in_channels=channel, out_channels=channel//reduction, kernel_size=1, stride=1, bias=False)

    self.relu = nn.ReLU()
    self.bn = nn.BatchNorm2d(channel//reduction)

    self.F_h = nn.Conv2d(in_channels=channel//reduction, out_channels=channel, kernel_size=1, stride=1, bias=False)
    self.F_w = nn.Conv2d(in_channels=channel//reduction, out_channels=channel, kernel_size=1, stride=1, bias=False)

    self.sigmoid_h = nn.Sigmoid()
    self.sigmoid_w = nn.Sigmoid()

def forward(self, x):

    x_h = self.avg_pool_x(x).permute(0, 1, 3, 2)
    x_w = self.avg_pool_y(x)

    x_cat_conv_relu = self.relu(self.conv_1x1(torch.cat((x_h, x_w), 3)))

    x_cat_conv_split_h, x_cat_conv_split_w = x_cat_conv_relu.split([self.h, self.w], 3)

    s_h = self.sigmoid_h(self.F_h(x_cat_conv_split_h.permute(0, 1, 3, 2)))
    s_w = self.sigmoid_w(self.F_w(x_cat_conv_split_w))

    out = x * s_h.expand_as(x) * s_w.expand_as(x)

    return out

if name == ‘main’:
x = torch.randn(1, 16, 128, 64) # b, c, h, w
ca_model = CA_Block(channel=16, h=128, w=64)
y = ca_model(x)
print(y.shape)


 3、ECA模块


论文：[(PDF) ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks (researchgate.net)]( ) ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks (researchgate.net)")


参考：[注意力机制（SE、Coordinate Attention、CBAM、ECA，SimAM）、即插即用的模块整理\_吴大炮的博客-优快云博客\_se注意力机制]( )


![](https://img-blog.csdnimg.cn/62ead8be56f74536b191476a5532d431.png)


ECA首先通过全局平均池化，然后利用卷积进行特征提取，实现跨通道的交互。


4、CBAM模块


论文： [[1807.06521] CBAM：卷积块注意模块 (arxiv.org)]( )")


参考：[注意力机制之《CBAM: Convolutional Block Attention Module》论文阅读\_落樱弥城的博客-优快云博客]( )


![](https://img-blog.csdnimg.cn/9c27ae909c724745800680a4e277220d.png)


![](https://img-blog.csdnimg.cn/3d800630460c4cb0abf0da5feb3ee088.png) 


CBAM模块分为channel-wise attention和spatial attention，通道注意力和SE结构相同，只是加了一个maxpooling，中间共享一个MLP，最后将两部分的输出相加经过sigmoid。


空间注意力使用平均池化和最大池化对输入特征层进行通道压缩，在使用卷积操作。

class ChannelAttention(nn.Module):
def init(self, in_planes, ratio=16):
super(ChannelAttention, self).init()
self.avg_pool = nn.AdaptiveAvgPool2d(1)
self.max_pool = nn.AdaptiveMaxPool2d(1)

    self.fc1   = nn.Conv2d(in_planes, in_planes // ratio, 1, bias=False)
    self.relu1 = nn.ReLU()
    self.fc2   = nn.Conv2d(in_planes // ratio, in_planes, 1, bias=False)

    self.sigmoid = nn.Sigmoid()

def forward(self, x):
    avg_out = self.fc2(self.relu1(self.fc1(self.avg_pool(x))))
    max_out = self.fc2(self.relu1(self.fc1(self.max_pool(x))))
    out = avg_out + max_out
    return self.sigmoid(out)

class SpatialAttention(nn.Module):
def init(self, kernel_size=7):
super(SpatialAttention, self).init()

    assert kernel_size in (3, 7), 'kernel size must be 3 or 7'
    padding = 3 if kernel_size == 7 else 1

    self.conv1 = nn.Conv2d(2, 1, kernel_size, padding=padding, bias=False)
    self.sigmoid = nn.Sigmoid()

def forward(self, x):
    avg_out = torch.mean(x, dim=1, keepdim=True)
    max_out, _ = torch.max(x, dim=1, keepdim=True)
    x = torch.cat([avg_out, max_out], dim=1)
    x = self.conv1(x)
    return self.sigmoid(x)


5、GAM模块


论文：[https://paperswithcode.com/paper/global-attention-mechanism-retain-information]( )


GAM注意力机制分为两个模块：CAM和SAM，通道注意是学习不同通道的权值，并用权值对不同通道进行多重划分，空间注意关注目标在图像上的位置信息，并通过空间特征的加权选择性的聚焦每个空间的特征。


![](https://img-blog.csdnimg.cn/986e9331c4074bdc8c078fe6ee7ab33e.png)


 通道注意力模块首先重新排列图像三维信息，然后通过MLP来放大跨维通道空间，如图6所示。在空间注意子模块中，使用两个卷积层进行空间信息融合，如图7所示，这样使通道更能关注空间信息。


![](https://img-blog.csdnimg.cn/6459baf4a47f46c2b59c0bb9494f50df.png)

import torch.nn as nn
import torch

class GAM_Attention(nn.Module):
def init(self, in_channels, out_channels, rate=4):
super(GAM_Attention, self).init()

    self.channel_attention = nn.Sequential(
        nn.Linear(in_channels, int(in_channels / rate)),
        nn.ReLU(inplace=True),
        nn.Linear(int(in_channels / rate), in_channels)
    )

    self.spatial_attention = nn.Sequential(
        nn.Conv2d(in_channels, int(in_channels / rate), kernel_size=7, padding=3),
        nn.BatchNorm2d(int(in_channels / rate)),
        nn.ReLU(inplace=True),