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

请问一下哪里可以获得全脸的9个AU group 区域呢? #21

Open
Gepeng18 opened this issue Dec 13, 2020 · 8 comments
Open

请问一下哪里可以获得全脸的9个AU group 区域呢? #21

Gepeng18 opened this issue Dec 13, 2020 · 8 comments

Comments

@Gepeng18
Copy link

您好,大牛

  1. 我看到您的AUDataset 的get_example() 输出的结果中有bbox,这个是9个AU group 区域嘛?
  2. 请问 https://github.com/machanic/AU_R-CNN/blob/master/AU_rcnn/datasets/AU_dataset.py#L138-L202 ,是将 AU_box_dict 转化为最终的9个AU group的bbox输出的代码嘛?
    谢谢大牛
@machanic
Copy link
Owner

@Gepeng18 最主要是FaceMaskCropper.get_cropface_and_box, 其实在config.py里面定义了精细的区域划分规则(利用了哪几个landmark point)以及AU group的规则。

@Gepeng18
Copy link
Author

@machanic 谢谢大牛的回复,小弟菜,不是很能弄明白,再研究研究。 PS: 期待大牛的Pytorch版本的代码。

@machanic
Copy link
Owner

@Gepeng18 我做好后会通知你的,其实现在有个半成品,但是因为在忙另一个会议,你要的话半成品的pytorch代码可以发你。

@Gepeng18
Copy link
Author

大牛,不好意思,才看到。 如果您方便的话,可以把Pytorch半成品发给我吗? 非常感谢。 还有,想请教您一个问题,您论文中关于ConvLSTM扩展的时候,是将9个AU组对应的特征图序列,每个特征图序列都送入一个ConvLSTM,所以其实是有9个ConvLSTM吗? 然后ConvLSTM后面的两个FC层,是这9个输出共用FC层,还是说每个输出都送入自己的FC层,一共有18个FC呢?
谢谢大牛解答,非常感谢

@Gepeng18
Copy link
Author

大牛,我的email : [email protected]

@machanic
Copy link
Owner

@Gepeng18 已经发了

@pakchoi-php
Copy link

你好,我想参考一下AU选区那部分,不知道是否能获得一份pytorch版本代码,这是我的邮箱[email protected]

@AMG024
Copy link

AMG024 commented Apr 15, 2022

大佬,我也想借鉴一下AU选取那块,请问不否可以把pytorch代码分享给我一份(半成品也行啊,我主要是想看一下区域划分那部分是怎么定义的,挺急的),万分感谢。这是我的邮箱[email protected]

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