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

阅读论文有几个疑问? #7

Open
ZHIZIHUABU opened this issue Aug 30, 2023 · 2 comments
Open

阅读论文有几个疑问? #7

ZHIZIHUABU opened this issue Aug 30, 2023 · 2 comments

Comments

@ZHIZIHUABU
Copy link

1、训练时所需要的fg_mask是在确定扩充异常样本的粘贴位置吗,避免缺陷粘贴在背景区域?
2、训练时分为两个阶段,分别是异常边界生成和边界引导,这两个训练过程是同步的吗?
3、选择的课件异常样本数量应该怎么选择呢,mvtec数据集默认每个类别选择10个异常样本,假设我有200个异常样本,3000个正常样本,异常样本数量选择多少呢?

@xcyao00
Copy link
Owner

xcyao00 commented Aug 30, 2023

  1. 是的,fg_mask用于提供前景区域,避免缺陷生成在背景区域。
  2. 是先生成边界再进行边界引导,但在代码中是在每个batch中动态确定边界。
  3. 异常样本应该是越多越好,看你总共有多少个异常样本可用,我们在mvtec上选择10个,是基于异常稀缺性的假设。

@ZHIZIHUABU
Copy link
Author

我现在想基于这种无监督模型去获取异常样本的位置,然后将位置坐标作为提示点输入到SAM进行分割,这样获取异常缺陷的mask,相当于分割标签了,因为直接用无监督模型训练得到的mask似乎是不太准确的。我最近看到了RSPrompter: Learning to Prompt for Remote Sensing Instance Segmentation based on Visual Foundation Model这篇论文似乎表达了类似的意思

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

2 participants