Simone Gasparini

> Hi, > > I have the same issue, but running on Ubuntu. When detection script is running on CPU everything is OK, detect the candidates, but when is running...

I think the problem might come from the missing architecture flags for Jetson Nano. It requires `arch=compute_53,code=sm_53` (see https://forums.developer.nvidia.com/t/jetson-nano-running-openpose-example-gives-a-cuda-check-failed/77196/3) but they are not in our list https://github.com/alicevision/CCTag/blob/develop/CMakeLists.txt#L166 You can try...

Memory could very well be the issue. You can try to crop that image around one of the cctag and process that one. A smaller image require less image. I...

I'm not an expert in cuda either. The cuda code was optimized to run in real time on GTXs. It is possible some of the configurations used for blocks, num...

thanks for testing that. I don't know if it is doable, but it would be nice if we could parametrize these parameters according to the architecture. It's nice to hear...

Just to know, are you using the cuda version or the cpu version? Do you have the same if u run it with `--use-cuda`?

I was asking because in the cuda version there are a lot of dynamic allocations but it should not be the case in the cpu. Can u post the log...

Thanks for this! We may need to better handle the c++17 flag because that also affects the cuda compiler and the minimum cuda version that is supported (11+). We have...

> @simogasp Could we maybe just use `std::mutex` and `std::lock_guard` to avoid conditional compilation? The project uses C++14 as a minimum, so `std::mutex` should perform just as well. yes that's...

Thanks for initiating this PR! The issue is now addressed in #200 so I'm closing this one.