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

c++ CPU mode how to use nms #247

Open
dbrazey opened this issue Jul 8, 2016 · 0 comments
Open

c++ CPU mode how to use nms #247

dbrazey opened this issue Jul 8, 2016 · 0 comments

Comments

@dbrazey
Copy link

dbrazey commented Jul 8, 2016

Hello,

I am writing because I have some troubles using py faster rcnn in c++.
In GPU mode, everything is ok : I can compile the library, learn using provided scripts and I can detect objects with my c++ code.

However, I am trying to use the library to detect objects only (no training) on a laptop without GPU card. In order to do this, I compile the library in CPU mode following the instructions found here :

#123

The file gpu_nms.so is not generated any more, I only get the cpu_nms.so file.
Then, when I compile my c++ code, I get the following error

/tmp/ccUNhddW.o : Dans la fonction « CFRCNN::DetectObjects(cv::Mat_) » :
FRCNN.cpp:(.text+0xcac) : référence indéfinie vers « nms(int, int_, float const_, int, int, float, int) »
collect2: error: ld returned 1 exit status

In GPU mode, the following c++ function call worked fine
#include "gpu_nms.hpp"
_nms(keep, &nbObjects_nms, sorted_pred_cls, nbRois, 5, m_nmsThreshold, 0);`

but the function _nms seems to be different in cpu_nms.so.
I didn't found any header like "cpu_nms.hpp" in the library.
How to use this function in CPU mode ?

Thank you for your help :)

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