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

About correlation layer. #72

Open
MeLonJ10 opened this issue Aug 24, 2021 · 1 comment
Open

About correlation layer. #72

MeLonJ10 opened this issue Aug 24, 2021 · 1 comment

Comments

@MeLonJ10
Copy link

MeLonJ10 commented Aug 24, 2021

Hi @AlessioTonioni, thanks for sharing your work!
I have a question about correlation layer defined in correlation_tf.
Does the correlation layer defined in this way mean that, it only use single pixel to calculate correlation rather than a patch centered at the candidate pixels?
Thank you so much!

@AlessioTonioni
Copy link
Member

Hi,
Yes the correlation layer computes correlation between single locations of the volumes provided as inputs.
In our architecture it is applied at lower scales in the netwrok and on the output of convolutional layers, therefore each location contains information coming from all the receptive field of the previous convolutional layers. For this reason the layer operates on single locations but it can be seemn as computing correlation between patches.

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

2 participants