yolov8s_pose.hef for NV12 compilation

General Discussion
1

Hello community,

first of all: I am a mechatronics engineer not a programmer (but I do my best to tackle the programming wit AI support :slight_smile: )

What I want to setup:

I have a Pi 5 with an Hailo8 26top ai hat. I want to setup a fast and light yolov8s_pose model.

Issues:

(maybe someone can compile a yolov8s_pose.hef for a pi5 with hailo8 ai hat with NV12 vid input)

the raspberry module3 camera sends the vid stream via NV12, but the pre-compiled *.hef models only support RGB input. I wanted to compile a yolov8s_pose.hef with NV12 to RGB conersion on the hailo8 chip. This where the diaster started :smiley:

Docker for windows = no nvidia support

ubuntu native = no nvidia support and NV12 activation did not work

I am a bit desperate currently…

2

Hi @user793 ,

Maybe this would be helpful: hailo-apps/hailo_apps/python/pipeline_apps/pose_estimation at main · hailo-ai/hailo-apps · GitHub

Thanks,

3

yes, in general I agree, but I was searching for the optimal solution for ~60fps detection. This is why I concluded to the version which is directly compiled for the hailo8 (the yolov8s_pose.hef) as this should show a better performance (?)

4

Hi,

The models used there are compiled for Hailo 8, 8L & 10H.

Thanks,