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

【论文阅读】Further Non-local and Channel Attention Networks for Vehicle Re-identification
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. ????
【论文阅读】Further Non-local and Channel Attention Networks for Vehicle Re-identification

The spatial attention block (SAB)

【论文阅读】Further Non-local and Channel Attention Networks for Vehicle Re-identification
正常的non-local结构
【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
通道注意力结构

【论文阅读】Further Non-local and Channel Attention Networks for Vehicle Re-identification
SAB的作用不大

相关文章:

  • 2021-12-22
  • 2021-04-19
  • 2022-01-10
  • 2021-08-17
  • 2021-04-01
  • 2021-11-06
  • 2021-06-14
  • 2021-11-01
猜你喜欢
  • 2021-05-08
  • 2021-11-11
  • 2021-06-06
  • 2021-07-19
  • 2022-01-21
  • 2021-11-10
  • 2022-01-21
相关资源
相似解决方案