-
Notifications
You must be signed in to change notification settings - Fork 751
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
no jniopencv_core and gomp in path on Jetson TX2 #769
Comments
Duplicate of #735 |
I'll let @kwatters answer that one. :) |
I haven't built opencv with gpu support yet ...but.. I do have the arm64 jars for javacv / opencv 1.5 here. These aren't official builds, but they do work with the Nano. I have a feeling that cuda & opencv is a bit of a deeper topic... as far as I can tell, Intel GPUs are pretty well supported in OpenCV.. but I can't say the same for Cuda. Anyway, grab the javacv arm64 jars there and you'll get the dependency you are looking for to get opencv / javacv working on the arm64 ( jetson ) boards. |
Ah, I see. We'll need to add |
@kwatters You are right that Cuda and opencv are a bit more of a deep topic, but to do it on a Intel GPu beats the advantage of cuda. What needs to be done is to allow to choose the cuda version. It will be more work but it may be easier to support the Jetson line in a whole. Not just the Jetson Nano. @saudet I will try to compile it with those things added and an other line since it does not exist on the cuda-100 branch. I will report back when it's done compiling |
@saudet You sir, are a genius. That was the problem. When I added that single bit it just launched. The only problem now is that it won't find opencv_cuda. In the screenshot below everthing with cuda the error is:
Below is the output from loading the java app with opencv
when the cmake command finished it did say that cuda was being used. What could be wrong now? At this point |
|
Is it just a case of removing the places where it loads opencv_cuda or do I need to do some more things?
|
We can remove the |
Ok, thanks for the help. Then I will open a issue on the opencv repo. |
Unless you plan to implement |
…inux-arm64` and `linux-ppc64le` (issue #769) * Upgrade presets for OpenCV 4.1.1, NumPy 1.17.0, nGraph 0.24.0
I finaly got cuda and opencv to compile on my Jetson TX2 but when I run it the program says it cannot find jniopencv_core and gomp but they are in the jar file opencv-4.0.1-1.5-linux-arm64-gpu.jar
What am I doing wrong?
The text was updated successfully, but these errors were encountered: