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

区域特征提取 #24

Closed
HN123-123 opened this issue Nov 26, 2021 · 2 comments
Closed

区域特征提取 #24

HN123-123 opened this issue Nov 26, 2021 · 2 comments

Comments

@HN123-123
Copy link

我尝试使用您的代码自己常见特征集,但是还有部分疑问。region_after和region_before 分别取自proposal_box_features和proposal_box_features1(均值),这有何区别??
另外我去https://github.com/facebookresearch/grid-feats-vqa下也并未找到相应区域yaml文件,请问您是否可以分享下您的X-101-region.yaml和X-152-region.yaml文件??

@HN123-123
Copy link
Author

另外,我发现您提供的区域特征生成文件直接将特征保存为.hdf5文件形式,我如果想保存源形式的话应该怎么做呢??

@luo3300612
Copy link
Owner

你好,这两者的区别是,before是roi pooling之后的特征,after是before再经过检测头的特征,我们用的模型是faster rcnn dc5,如果没有记错的话,after就是before经过fc之后的特征,在我们的实验中after的结果没有before好,所以我们用的就是before的特征

主要的数据都在这三行,你把data=后面的tensor拿出来单独保存就行了

如果你想保存成npy的话,就把feature box size在一起保存成npz
如果你想保存成pth的话,可以弄一个字典用torch.save保存成pth

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