Live Inference using YoloV8

[PiyushKamboj1503]

Search before asking

• I have searched the Supervision issues and found no similar feature requests.

Question

The argument show=True give the live inference in YoloV8 with it own bounding boxes.
How to get the live prediction in a window in similar way using supervision with the custom bounding boxes and zones.

Additional

No response

[SkalskiP]

Hi 👋🏻! Given the fact that it is not an issue related to the library itself, I'll move that conversation to our Q&A discussions

[SkalskiP]

1. Use get_video_frames_generator
2. Run detection with YOLOv8 on the current frame
3. Convert YOLOv8 result to Detections object
4. Annotate the frame with BoxAnnotator.
5. Use OpenCV imshow to display annotated images.

Here is a pseudo-code possibly would require some fixing but could be useful as the first iteration.

import cv2
import supervision as sv
from ultralytics import YOLO

model = YOLO('yolov8s.pt')
annotator = BoxAnnotator()

for frame in sv.get_video_frames_generator(source_path='source_video.mp4'):
    yolov8_results = model(frame)[0]
    detections = sv.Detections.from_yolov8(yolov8_results)
    annotated_frame = annotator.annotate(scene=frame.copy(), detections=detections)
    cv2.imshow('image window', annotated_frame)