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

模型转化后,有一部分小物体的检测出现框明显变小一圈的情况。 #66

Open
MichaelSunEngineer opened this issue Feb 27, 2020 · 1 comment

Comments

@MichaelSunEngineer
Copy link

MichaelSunEngineer commented Feb 27, 2020

@ChenYingpeng
[upsample] stride=2
[route] layers = -1, 36
检测小物体时候作者提出这样的修改。
修改模型,并且剪枝,然后训练得到weights文件。在darknet下一切正常,当转化为caffemodel以后,出现问题。锚点我也对应自己的数据集修改过了,和训练时一样。会不会是因为改了层,锚点对应发生错误,具体该怎么修改呢?

@MichaelSunEngineer
Copy link
Author

@ChenYingpeng 原来的cfg参数是
[upsample] stride=4
[route] layers = -1, 11
我修改之后,一张图片里大点的物体和特别小的物体包围框都是准确的,可是中间的物体包围框会特别小,不能吧这个物体完全包围。等这个物体走近变大的时候,包围框就又成了准确的。且这个模型是我剪枝过2次以后的。希望大神能给我解答下这个问题。

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

1 participant