Aruco marker detection with Nvidia Jetson TX2

Hello everyone .I have the Nvidia Jetson TX2 and I would like to use the camera as to do aruco detection and camera pose estimation.I installed OpenCV by following this video ( library is not coming.I would like to ask about how to install it and if needs something more (means camera of jetson) as to do that.I would like to mention that I have done the camera pose estimation and detection with my laptop and it works perfectly.

Thanks for the consideration,

George Daskalakis

Hi George,

I cannot tell what is going on because you didn’t share any error with us.

Thus, some possibilities for you

  1. Please make sure you are using the opencv library built from your own source but not from any other opencv (debian package or from jetpack).

  2. Please make sure opencv_contrib/modules/aruco is really configured during cmake.

Thank you Wayne for your reply.I would like to say you that I installed Opencv on my own but if I check in the terminal the libraries aruco is not there.I thought that as in the pc will come automatically.I have to adjust that with a way as you mentioned also.Can you suggest me methods as to do that.?

George Daskalakis

Hi George,

I am not sure what is going on here since I don’t actively work on OpenCV. If you said you could see it on PC, could you compare the installation steps between pc and tegra?

Do you use same configuration?

Basically I follow with Jetson this installation of opencv ( and for Pc ( Jetson I face error with Cmake and the confiduaration was not completed,so that’s why i followed an other installation.

when you enable opencv_contrib - it will build modules including , but not limited to the aruco.
However, I would rather look into sourceforge cpp aruco sources and additionally python-aruco package rather than the default opencv_contrib aruco pack.

If you want to use opencv with aruco module then either you need to pip install opencv contrib module or build opencv with opencv contrib. Follow this link and you can use it with any jetson module: