当前位置:网站首页>【论文阅读】Further Non-local and Channel Attention Networks for Vehicle Re-identification
【论文阅读】Further Non-local and Channel Attention Networks for Vehicle Re-identification
2022-08-04 05:29:00 【≈落小朵】
问题: 类间差异小,类内差异大
提出:双分支自适应注意网络
在视觉皮层双流理论的启发下, 基于non-local和channel关系 ,构建了一个双分支FNC网络来捕获多种有用信息
(消除背景的影响)
Further Non-local and Channel attention (FNC) is constructed to simulate two-stream theory of visual cortex
提出了一种有效的 注意力融合方法 ,充分模拟了空间注意力和信道注意力的影响。
Proposed method
Then, we change the last spatial down-sampling operation stride from 2 to 1 to provide a large spatial view for the spatial attention module, thereby capturing highly detailed spatial correlations. ????
The spatial attention block (SAB)
正常的non-local结构
中间是不是少写了一个公式,中间的相似图和T相乘的没有写,后面的 x? 是什么意思?
- 正常的non-local结构是直接相加在一起,但是这篇文章选择采用sigmoid函数激活(是可以获得更加显著的特征么)
利用sigmoid函数,可以增加权值对特征图的影响,并引入非线性因素(这部分-----)
通道注意力结构
SAB的作用不大
Then, we change the last spatial down-sampling operation stride from 2 to 1 to provide a large spatial view for the spatial attention module, thereby capturing highly detailed spatial correlations. 调整下采样的步长
边栏推荐
- sql中group by的用法
- SQL练习 2022/7/4
- 光条中心提取方法总结(二)
- Data reading in yolov3 (1)
- TensorFlow2学习笔记:4、第一个神经网模型,鸢尾花分类
- sklearn中的学习曲线learning_curve函数
- Endnote编辑参考文献
- Pytorch语义分割理解
- [Deep Learning 21 Days Learning Challenge] Memo: What does our neural network model look like? - detailed explanation of model.summary()
- MFC读取点云,只能正常显示第一个,显示后面时报错
猜你喜欢
随机推荐
[Deep Learning 21 Days Learning Challenge] Memo: What does our neural network model look like? - detailed explanation of model.summary()
动手学深度学习_softmax回归
yolov3数据读入(二)
图像形变(插值方法)
Androd Day02
线性回归02---波士顿房价预测
Data reading in yolov3 (1)
Usage of RecyclerView
PyTorch
剑指 Offer 2022/7/12
【CV-Learning】Object Detection & Instance Segmentation
【深度学习21天学习挑战赛】备忘篇:我们的神经网模型到底长啥样?——model.summary()详解
Various commands such as creating a new user in postgresql
Qt日常学习
Endnote编辑参考文献
软著撰写注意事项
安装dlib踩坑记录,报错:WARNING: pip is configured with locations that require TLS/SSL
postgres recursive query
PostgreSQL模式(Schema)
Deep Adversarial Decomposition: A Unified Framework for Separating Superimposed Images