How to natively install ODM with GPU accelerate?

Hi,
I natively install ODM, while, when I use my natively ODM, the gpu useage did not seem increased, so, I want to know the method how to natively install ODM with gpu accerlate which support cuda.

Thanks a lot!!

1 Like

Welcome!

Really good question…

Have you looked at how the GPU-enabled Docker image is being built/modified?
https://hub.docker.com/layers/opendronemap/odm/gpu/images/sha256-899037cc8bfbdc6119c755d34136d64a50afb86ea7ce80f9e0747adcf94a74dd?context=explore

2 Likes

Thanks a lot for your reply!
I see the link, it seems like a docker configuration list.
And I still do not know how to make it. The cuda version I have is 11.1, not as the link above 11.2.
I’m wondering if I could check a cmake label, something like CUDA or GPU, I have checked the source code, and I saw some .cpp files have the name with “cuda”.

It looks like a ton of environment variables are being set as well.