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

Question about the "share_planes" #66

Open
BuLingBin opened this issue Nov 4, 2022 · 0 comments
Open

Question about the "share_planes" #66

BuLingBin opened this issue Nov 4, 2022 · 0 comments

Comments

@BuLingBin
Copy link

Hi, it is a nice work!
But I am confused about the "share_planes" in PointTransformerLayer.
n, nsample, c = x_v.shape; s = self.share_planes
x = ((x_v + p_r).view(n, nsample, s, c // s) * w.unsqueeze(2)).sum(1).view(n, c)
Apparently, w's dimension is reduced by Linear, which is not illustrated in the paper. I think this operation is not consistent with the vector attention, It is more like a compromise of scalar attention and vector attention.
Why partition the feature dimension of (x_v + p_r) into share_planes?

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