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

Mask filtering Script #31

Open
zhh6425 opened this issue Dec 20, 2024 · 2 comments
Open

Mask filtering Script #31

zhh6425 opened this issue Dec 20, 2024 · 2 comments

Comments

@zhh6425
Copy link

zhh6425 commented Dec 20, 2024

Hi, Thanks for the great work!!

I have noticed the paper mentions a set of mask filtering processes for the MPM module. But I can't find the code related to that. How can I generate masks on my own datasets with the same mask filtering in the paper?

And also, will you provide the mask3d checkpoint with Mask scoring that you mention in the paper?

Thanks!

@zhuoyuan00
Copy link

Yes I did not find the PMP code either. It seems the authors provided the masks directly.

@ZheningHuang
Copy link
Member

Dear Both,

Apologies for the delayed response—I was not very active due to submissions and being on annual leave.

The Mask Proposal Module (MPM) was inherently adapted from Mask3D. In the demo here, MPM is included to produce mask for our input pcd.

Regarding the benchmark results, to ensure a fair comparison with other works and ease of use—since many methods at the time of OpenIns3D followed similar strategies for mask proposal—we provided mask proposals directly.

If you want to generate masks organically, I recommend using Mask3D easy setup, as the mask results were originally obtained from there. Additionally, Segment3D appears to be a stronger version of the class-agnostic mask proposal.

Zhening

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