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

Questions about Detail Emphasis Module #8

Open
Shaw-Way opened this issue Nov 8, 2023 · 0 comments
Open

Questions about Detail Emphasis Module #8

Shaw-Way opened this issue Nov 8, 2023 · 0 comments

Comments

@Shaw-Way
Copy link

Shaw-Way commented Nov 8, 2023

Hi, thanks for your remarkable work. I have some questions about Detail Emphasis Module.

class DEM(nn.Module):
def __init__(self, channel):
""" Detail Emphasis Module """
super(DEM, self).__init__()
self.conv1 = nn.Sequential(nn.ReflectionPad2d(1),
nn.Conv2d(channel, channel, kernel_size=3, stride=1, padding=0),
nn.BatchNorm2d(channel),
nn.ReLU(True))
self.global_path = nn.Sequential(nn.AdaptiveAvgPool2d(1),
nn.Conv2d(channel, channel, kernel_size=1, stride=1, padding=0),
nn.ReLU(True),
nn.Conv2d(channel, channel, kernel_size=1, stride=1, padding=0),
nn.Sigmoid())

I noticed that it contains 2 parts, the first is conv1, the second is global_path. It seems the second one is similar to Channel Attention Module. But I have no idea what is the function of the first part(conv1). May I ask how you conceived the first part or where it came from? Thanks in advance.

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

1 participant