Jetson Orion NX 8GB

Autonomous Machines Jetson & Embedded Systems Jetson Orin NX
1

Hi, I have created a Yolov8-Large model that does inferencing on 4 realtime cctv footage but the speed of prediction/infrencing is very slow like 10fps and i have created a python based model without onxx .

Please guide me with some tips how to make model do predictions faster on realtime video at 30 fps or more with scalability .

Regards
Rahul Buttan

3

Hi,

Could you run the tegrastats to check if GPU resources are fully used?

$ sudo tegrastats

Thanks.

5

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.