I’m writing Computer vision based guidance and it seems like the processing time is high on real drone vs simulation.
Is the simulation code running at same performance than real drone, or is it using local computer performances ?
Also, what are the specs of Anafi AI CPU ? I can’t find the information …
It’s been a month without any help.
I have the opencv cv::aruco::detectMarkers function that runs in around 2ms with Sphinx, and takes around 200ms on the real drone. It runs even faster on a raspberry …
Can we know more about the CPU of Anafi AI ?
Are you limiting the CPU to external missions that are not parrot ? The huge processing time of this very fast function is weird, especially on a 5k€ drone.