If you are looking for minimal working example to run multiple pipelines on multiple Hailo-8 chips, using multiple neural networks, here is a minimal working example
(MWE). It is a step-by-step guide on how to set up multi-stream inferencing pipeline on FALCON-H8 from Lanner Electronics Inc..
In this example, there are 4 Hailo-8 chips on one PCIe card.
One chip processes one live camera stream. As I have only 3 cameras, the last chip can be used to process a recorded video clip.
One neural network is assigned to a specific chip. I used YOLOv8m and YOLOv7. Feel free to try other neural networks.
Codes are simple, intuitive, and modular. This MWE is based on the Hailo application example codes, but tailored for multi-chip device like FALCON.
If you like to add more features, feel free to comment or contribute.