YoloV8 cant recover memory Python

Hey,

Any chance anyone has any idea how I can recover the memory consumed by my model when it inferences by cpu. I have traced the memory usage and see where it is used but I cant seem to recover it. See below:
I am only processing one image every 2 mins but need to recover the memory as I am limited to 4GB on a Pi

I’ve tried a few ways to force a clean up but no luck

Any help would be appreciated.

PS I am only interested in the bounding boxes if that helps

Hi! We are moving the Ultralytics community to Discord. To receive support for your questions, problems, issues, etc. please join us on the new server: Ultralytics