YOLOv8 Object Counting in Real-time with Webcam, OpenCV and Supervision

Curious about how to build an application capable of detecting objects on a camera stream in real time? You are in the right place! Together we will learn how to access the webcam using OpenCV, detect objects with YOLOv8, and count them with Supervision.

Chapters:

0:00 Introduction
0:38 Webcam inference with YOLOv8 CLI
02:12 Accessing webcam with OpenCV
07:05 Webcam inference with YOLOv8 SDK
08:25 Drawing bounding boxes
10:05 Using polygon zones to count detections
14:28 Eliminating double detections with agnostic NMS
16:07 Few extra tricks
18:59 Conclusion

Resources:

🌏 Roboflow: https://roboflow.com

🌌 Roboflow Universe: https://universe.roboflow.com

📷 Demo repository: https://github.com/SkalskiP/yolov8-live
🤩 Supervision documentation: https://roboflow.github.io/supervision
⭐ Supervision repository: https://github.com/roboflow/supervision

Stay updated with the projects I'm working on at https://github.com/roboflow and https://github.com/SkalskiP! ⭐