Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

论文用三个可学参数作为融合权重,但这3个参数如何学到? #86

Open
LDoubleZhi opened this issue May 12, 2020 · 2 comments

Comments

@LDoubleZhi
Copy link

论文中没有看到具体如何监督学习这3个参数,希望作者解惑

@zhenghao977
Copy link

zhenghao977 commented May 28, 2020

在ASFF模块中有,3个尺度通过插值和downsample到同一size后,每一个feature map经由一个1✖️1的卷积得到8维weights,然后3个尺度cat后再统一经过一个1✖️1的卷积得到3个参数,softmax后与3个feature map相乘得到fused feature map~ 因为这个fused feature map参与了loss计算 ,所以反向传播会更新相应的卷积层weights也就是学习到了这3个参数~

@joe660
Copy link

joe660 commented Dec 14, 2020

得到fused feature map~

说的太好了!我是按照asff在YOLO一个图理解的
image
1、表示cat 2、表示softmax 3、表示softmax后与3个feature map相乘得到fused feature map 是这个意思吗

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants