Click here to Skip to main content
15,066,299 members
Please Sign up or sign in to vote.
0.00/5 (No votes)
See more:
I have a cv::KeyPoint class in the caller function in CPU. I wanted to pass its vector as reference to a kernel function of CUDA. How can I access its members in a kernel function of CUDA? I am trying to implement this on CUDA 6.0.

I Googled for its solution, but didn't succeed. Please help.

Thanks in advance.
Posted
Updated 8-Oct-14 21:16pm
v4
Comments
Richard MacCutchan 8-Oct-14 9:02am
   
You have edited your post but have not added any useful information. stating that you want to do something is not a good question. You need to show some code and explain why it does not work. Have you tried the support forum that I suggested some days ago?
SanjaySMK 6-Nov-14 7:59am
   
Hey Richard, sorry for the delay in reply. Many thanks for the solution, the thing is I could have installed CUDA 6.5 which has Thrust library, which is based on STL. But because of the time constraint, I could not install it. So, I used float pointer to get the first address of that vector, and passed it to the kernel function, and that has worked. Later, when time permits, I will be surely installing CUDA 6.5.

Again, many many thanks for the solution.

1 solution

Try the support forum at http://www.nvidia.com/page/support.html[^].
   

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)




CodeProject, 20 Bay Street, 11th Floor Toronto, Ontario, Canada M5J 2N8 +1 (416) 849-8900