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

为什么,预测出来的边框比较宽? #91

Open
EricHuiK opened this issue May 23, 2021 · 6 comments
Open

为什么,预测出来的边框比较宽? #91

EricHuiK opened this issue May 23, 2021 · 6 comments

Comments

@EricHuiK
Copy link

No description provided.

@xinchengxxc
Copy link

same question. I tried many fancy networks, all with "thicker" edges comparing with gt. any ideas?

@stupidman98
Copy link

LPCB和DSCN减轻了这个问题,但是没有开源。原因包括损失函数和上采样时分辨率的丢失,我现在正在解决这个问题

@EricHuiK
Copy link
Author

EricHuiK commented Dec 7, 2021 via email

@xinchengxxc
Copy link

Towards improving edge quality using combinatorial optimization and a novel skeletonize algorithm
https://bmcmedimaging.biomedcentral.com/articles/10.1186/s12880-021-00650-z
我找到一篇边缘后处理的论文,用的方案很safe,基本上是parameter search 通过调整参数的后处理得到仅宽一个pixel的边缘。这里分享一下

@stupidman98
Copy link

Towards improving edge quality using combinatorial optimization and a novel skeletonize algorithm https://bmcmedimaging.biomedcentral.com/articles/10.1186/s12880-021-00650-z 我找到一篇边缘后处理的论文,用的方案很safe,基本上是parameter search 通过调整参数的后处理得到仅宽一个pixel的边缘。这里分享一下

谢谢你,这篇论文的评测指标或许对我很有帮助。我最近做了一个边缘检测模型能够很好的抑制噪声并产生大约3个像素宽的薄边缘,就是ods上不去,反应不了提升

@zhangxiao121
Copy link

对,我也尝试了一些方法,但是均衡了速度和效果[我这边实际场景对性能更加苛刻一些],所以使用了两阶段的方法,先检测位置,后分割

------------------ 原始邮件 ------------------ 发件人: "s9xie/hed" @.>; 发送时间: 2021年12月7日(星期二) 晚上6:40 @.>; @.@.>; 主题: Re: [s9xie/hed] 为什么,预测出来的边框比较宽? (#91) LPCB和DSCN减轻了这个问题,但是没有开源。原因包括损失函数和上采样时分辨率的丢失,我现在正在解决这个问题 — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

您好,你是先提取出边缘,然后再用语义分割的吗?

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

4 participants