Cupy and DNN_BACKENDCUDA -> All Fail on TX2

I cannot install or use any of these.

DNN_BACKENDCUDA

You need OpenCV 4.2 or above to use DNN_BACKEND_CUDA
Tried to install OpenCV after the jetpack installation using this guide → Says build successful but Python3 still states that the Jetpack’s Opencv 4.1.0 is the current version. Followed these steps verbatim except for Jetson 2 specific ARCH.
https://pyimagesearch.com/2020/02/03/how-to-use-opencvs-dnn-module-with-nvidia-gpus-cuda-and-cudnn/

Cupy
Stalls and fails completely.
Can’t upgrade Python 3.6.9 to anything else. Always states 3.6.9 is the default package install from Jetson and I can’t figure out how to delete 3.6.9 and keep 3.8+… Very frustrating
Cupy website states versions of Python supported: Python: v3.7.0+ / v3.8.0+ / v3.9.0+ / v3.10.0+

Hi,

Which JetPack version do you use?
We can install both on the latest JetPack 4.6.

OpenCV

cuPy

Please use compute_62 and sm_62 for TX2.
Thanks.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.