How to do object detection in C# with yolov8 weights?

See more:

C#

computer-vision

New to object detection in C#.

These were the steps to train my model.

!nvidia-smi
!pip install ultralytics
!pwd

import os
HOME = os.getcwd()
print(HOME)

from ultralytics import YOLO
from IPython.display import display, Image
from IPython import display
display.clear_output()
!yolo checks                               ... instead of !yolo mode=checks (warning error)



# roboflow copy project details
!pip install roboflow

from roboflow import Roboflow
rf = Roboflow(api_key="....")
project = rf.workspace("fcuv4h").project("test0xi")
version = project.version(1)
dataset = version.download("yolov8")



!yolo task=detect mode=train model=yolov8s.pt data={dataset.location}/data.yaml epochs=20 imgsz=640 

Now I downloaded the best.pt weight, but how should this custom weights be used in C#. Do these need to be converted to onnx format and what is best way to convert?

After if conversion is done what steps should be taken.

What any article or link to follow to complete inference in winform webcam c#.

What I have tried:

Already converted to onnx model but with help of chat gpt tried to inference but did not work.