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

Baseline code release ? #1

Open
trungpham2606 opened this issue Jun 30, 2021 · 16 comments
Open

Baseline code release ? #1

trungpham2606 opened this issue Jun 30, 2021 · 16 comments

Comments

@trungpham2606
Copy link

Dear @DIYer22
I just had read your paper. It's such a great work.
When will you release the baseline code for training and testing ?

@DIYer22
Copy link
Contributor

DIYer22 commented Jun 30, 2021

We are advancing open source work, the "building iShape" code is already released.
But the baseline code is still in progress. It should be released within two weeks at the latest.

Thanks for attention!

@trungpham2606
Copy link
Author

Oh, so glad to hear it from you.
Iam looking forward to testing it ^^

@trungpham2606
Copy link
Author

Dear @DIYer22
How is the release progress ^^

@DIYer22
Copy link
Contributor

DIYer22 commented Jul 20, 2021

@trungpham2606 The progress was interrupted by preparing rebuttals for NIPS and other things. We plan to advance open-source baseline code after 07/24. Sorry for the delay. We will notify you as soon as we release the baseline code.

@trungpham2606
Copy link
Author

Thank you for your prompt response @DIYer22

@trungpham2606
Copy link
Author

Hello, is everything oke @DIYer22 ?

@DIYer22
Copy link
Contributor

DIYer22 commented Aug 26, 2021

We try to allocate more time to push the progress, Really really sorry for the delay again!

@qianyizhang
Copy link

just a quick question, how is the irregular shape kernel implemented?
did you write new cuda ops or using F.gird_sample?

@DIYer22
Copy link
Contributor

DIYer22 commented Oct 14, 2021

@qianyizhang No, just by numpy. kernel is used on GT generation and build graph. No need to do it on the GPU

@qianyizhang
Copy link

@qianyizhang No, just by numpy. kernel is used on GT generation and build graph. No need to do it on the GPU

  1. when calculcating GT, since its not regular grid, how did you effectively aggregate the result? or you simply does a big for loop then stack?
  2. in training/inference, do you simply make a vector prediction which is supervised by the GT affinity? I thought you make aggregating of feature (by kernel) then calculate their pair-wise similarity/distance, which is then supervised by GT

@robbyneven
Copy link

Any ETA on baseline code release yet?

@invincible-28
Copy link

When will the baseline code be released? Please let us know.

@SEUZTh
Copy link

SEUZTh commented Mar 7, 2022

Any ETA on baseline code release yet?

@mchaniotakis
Copy link

Thank you for creating this segmentation method, could you please update us on the status of the code release? Thank you.

@xiongsheng001
Copy link

It's been a long time since you promised to open the source code.

@Asthestarsfalll
Copy link

Have you accidentally deleted your code?

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

9 participants