通道注意力：Squeeze-and-Excitation(SE) Attention

Brief introduction:
这是CVPR2018的一篇文章，同样非常具有影响力，目前引用量7k+。本文是做通道注意力的，因其简单的结构和有效性，将通道注意力掀起了一波小高潮。大道至简，这篇文章的思想可以说非常简单，首先将spatial维度进行AdaptiveAvgPool，然后通过两个FC学习到通道注意力，并用Sigmoid进行归一化得到Channel Attention Map,最后将Channel Attention Map与原特征相乘，就得到了加权后的特征。

import numpy as np
import torch
from torch import nn
from torch.nn import init



class SEAttention(nn.Module):

    def __init__(self, channel=512,reduction=16):
        super().__init__()
        self.avg_pool = nn.AdaptiveAvgPool2d(1)
        self.fc = nn.Sequential(
            nn.Linear(channel, channel // reduction, bias=False),
            nn.ReLU(inplace=True),
            nn.Linear(channel // reduction, channel, bias=False),
            nn.Sigmoid()
        )


    def init_weights(self):
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                init.kaiming_normal_(m.weight, mode='fan_out')
                if m.bias is not None:
                    init.constant_(m.bias, 0)
            elif isinstance(m, nn.BatchNorm2d):
                init.constant_(m.weight, 1)
                init.constant_(m.bias, 0)
            elif isinstance(m, nn.Linear):
                init.normal_(m.weight, std=0.001)
                if m.bias is not None:
                    init.constant_(m.bias, 0)

    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)
        print(y.shape)
        return x * y.expand_as(x)

if __name__ == '__main__':
    input=torch.randn(50,512,7,7)
    se = SEAttention(channel=512,reduction=8)
    output=se(input)
    print(output.shape)

    

地址：https://arxiv.org/pdf/1709.01507.pdf