DepthAI viewer yolo models resolutions OAK-1 POE

AugusteCouecou

Hello

I have a OAK-1 POE camera and my goal is to track people in an area so the resolution is pretty important.

I am pretty new to the world of Detection model embedded in a camera but I would like to know why when I run the depthAI viewer I have yoloV8 running perfectly smooth in the native resolution of the camera (1920x1080) and when I use the python sdk I can only use the 416x416 resolution and the yolo models seems to be trained at a max fixed resolution of 640x640.

If you have any pointers it would help me a lot

Thanks by advance

jakaskerl

AugusteCouecou
The models used are the same, the input resolution also. The difference is that the viewer upscales the NN results from 416x416 back to 1080p image.

AugusteCouecou

Thank you very much I will look into it 🙂