Hi All
We are currently working on a project using YOLO26. At the moment, we are using the standard model. We would like to understand its compatibility with commercially available edge devices or outdoor camera units.
What I am trying to ask is: we do not want to develop any hardware at this stage. Instead, we are looking for existing, off-the-shelf solutions where we can simply load our code and get the application running.
YOLO runs on many edge devices. It would work as long as you have compatible hardware.
At the moment, I am running a Raspberry Pi, but I am looking for something smaller. In the worst case, I could use a Raspberry Pi Zero 2 W, but I was wondering what would happen if I ran my models directly on the Pi itself alongside the Sony IMX500 camera. Would this cause a conflict, or could it be implemented as a two-stage processing system?
Pi Zero can run YOLO, but not in realtime. It will take several seconds to return a result for each frame.
They would run, but each model would return detections independently of the other. It’s like running two models at the same time. If you’re using IMX500, you run the model on camera by exporting to IMX500. The Pi doesn’t run the model. It processes the detections.