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

请问什么时候可以提供训练代码呀 十分感谢! #10

Open
WJJLBJ opened this issue Apr 17, 2024 · 10 comments
Open

请问什么时候可以提供训练代码呀 十分感谢! #10

WJJLBJ opened this issue Apr 17, 2024 · 10 comments

Comments

@WJJLBJ
Copy link

WJJLBJ commented Apr 17, 2024

我把loftr的backbone换成repvgg 其他的没有改动 发现结果非常差 请问有什么好的建议吗

@chicleee
Copy link

+1

1 similar comment
@YangHai-1218
Copy link

+1

@lichenyi112
Copy link

+1

@Adolfhill
Copy link

+10086

@wyf2020
Copy link
Contributor

wyf2020 commented Apr 23, 2024

非常感谢对我们工作的支持和关注!我们的训练代码还在整理中,预计在5月开源。另外,如果想把loftr的backbone换成repvgg后用loftr的代码训练,需要注意backbone和FPN的interpolate的align_corner=True和False要对应不同的supervision计算方式(可以参考这篇文章)。同时如果直接替换成repvgg会导致较小的精度损失,需要同时使用paper里3.4节的efficient fine feature extraction策略。

@lichenyi112
Copy link

+1

@r2race
Copy link

r2race commented Jun 11, 2024

+1+1+1 ... +999

@cjq216
Copy link

cjq216 commented Jul 6, 2024

+1

@hiyyg
Copy link

hiyyg commented Jul 6, 2024

非常感谢对我们工作的支持和关注!我们的训练代码还在整理中,预计在5月开源。另外,如果想把loftr的backbone换成repvgg后用loftr的代码训练,需要注意backbone和FPN的interpolate的align_corner=True和False要对应不同的supervision计算方式(可以参考这篇文章)。同时如果直接替换成repvgg会导致较小的精度损失,需要同时使用paper里3.4节的efficient fine feature extraction策略。

?

@wyf2020
Copy link
Contributor

wyf2020 commented Sep 4, 2024

非常抱歉训练代码没有准时开源,我们的训练代码现在已经开源,欢迎使用!

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