Object Detection Using YOLOv8: Real-Time Detection Made Simple
Ashish Kasama

Object Detection Using YOLOv8: Real-Time Detection Made Simple

Curious how your webcam can identify people, pets, or even products in real time?
Let’s dive into the world of object detection using the latest YOLOv8 model—fast, accurate, and beginner-friendly.

 What You'll Learn

  • How object detection works with bounding boxes

  • Understanding confidence scores

  • Real-time detection with your webcam

  • Tools: Ultralytics YOLOv8, OpenCV

What Is Object Detection?

At its core, object detection does two things:

  1. Finds objects in images or video (like a cat, bottle, or person).

  2. Draws bounding boxes around them with a confidence score.

YOLO (You Only Look Once) models are popular because they’re super fast and efficient, even on a webcam.

Why YOLOv8?

  • Trained on large datasets (like COCO)

  • Real-time inference (can process 30+ FPS on decent hardware)

  • Easy Python SDK via Ultralytics

  • Supports custom training for your own use case

Tools You'll Use

  • Ultralytics YOLOv8 – the latest version of the YOLO family

  • OpenCV – to capture and display webcam feed

  • Python – the glue that makes it all work

Installation

pip install ultralytics opencv-python 

Sample Code – Detect Objects from Webcam


from ultralytics import YOLO
import cv2

# Load YOLOv8 pretrained model
model = YOLO("yolov8n.pt")  # 'n' stands for nano (lightweight version)

# Start video capture
cap = cv2.VideoCapture(0)  # 0 is usually the default webcam

while True:
    ret, frame = cap.read()
    if not ret:
        break

    # Run YOLO detection
    results = model(frame)[0]

    # Draw boxes
    for box in results.boxes:
        x1, y1, x2, y2 = map(int, box.xyxy[0])
        confidence = box.conf[0]
        class_id = int(box.cls[0])
        label = model.names[class_id]
        
        # Draw rectangle and label
        cv2.rectangle(frame, (x1, y1), (x2, y2), (0, 255, 0), 2)
        cv2.putText(frame, f"{label} {confidence:.2f}", (x1, y1 - 10),
                    cv2.FONT_HERSHEY_SIMPLEX, 0.6, (255, 0, 0), 2)

    cv2.imshow("YOLOv8 Detection", frame)

    if cv2.waitKey(1) & 0xFF == ord("q"):
        break

cap.release()
cv2.destroyAllWindows()

 

Use Cases

  • Detect people for surveillance or attendance

  • Spot pets in home automation systems

  • Recognize products in retail or inventory management

  • Use for smart doorbells or robot vision

Bonus Tip: Customize It

Train YOLOv8 on your own dataset to detect brand logos, machines, or rare animals.
Ultralytics provides simple commands to train custom models too!

Final Thoughts

Whether you're building a smart camera, a retail scanner, or just learning for fun—YOLOv8 is a solid way to get started with object detection.

Want to take it further? Add voice alerts, count objects, or track movement in real-time.