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

关于SFTPM图像异常分割算法 #3844

Open
Hold-on-li opened this issue Nov 8, 2024 · 5 comments
Open

关于SFTPM图像异常分割算法 #3844

Hold-on-li opened this issue Nov 8, 2024 · 5 comments
Assignees

Comments

@Hold-on-li
Copy link

问题描述 Please describe your issue

请问一下,SFTPM图像异常分割算法支持C++推理部署吗?

@liuhongen1234567
Copy link
Collaborator

liuhongen1234567 commented Nov 8, 2024

您好,应该是支持的,可以参考 PaddleX 文档 https://github.com/PaddlePaddle/PaddleX/blob/release/3.0-beta1/docs/pipeline_usage/tutorials/cv_pipelines/image_anomaly_detection.md 3. 开发集成/部署-服务化部署部分

@Kaze816
Copy link

Kaze816 commented Nov 13, 2024

配置文件中的tain.txt 中是什么样式呀? @liuhongen1234567 这不是只需要正样本么,难道也需要生成些0值的标签图像?

train_dataset:
type: Dataset
num_classes: 1
dataset_root: /mv_dataset/grid # You need to specify a specific category root in mvtec_ad dataset
train_path: mv_dataset/grid/train.txt
...

@liuhongen1234567
Copy link
Collaborator

liuhongen1234567 commented Nov 15, 2024

您好,可以从这个页面 https://github.com/PaddlePaddle/PaddleX/blob/release/3.0-beta1/docs/module_usage/tutorials/cv_modules/anomaly_detection.md 下载 mvtec_examples 数据集看一下

@liuhongen1234567
Copy link
Collaborator

image 可以参考这个介绍,对SFTPM算法的描述已经比较清楚了,训练过程中的样本应该是没有异常的样本。train.txt 的标签如下: image

@Kaze816
Copy link

Kaze816 commented Nov 15, 2024

原来不明白的是 train.txt 里面的内容,正常的监督分割训练,此处应该是每行一组 图像 掩模 的路径,可这正样本没有掩模呀?
您这可算解决了我的一大疑惑,谢谢! @liuhongen1234567

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