Computer Vision with CVZone

Detect faces, hands and body poses in real time on a Raspberry Pi with Python, OpenCV and CVZone

Watch the video

Overview

Computer vision lets a computer make sense of what a camera sees. In this course you’ll use CVZone, a friendly Python library built on top of OpenCV and Google’s MediaPipe, to detect faces, track hands and estimate body poses in real time on a Raspberry Pi. Each project is only a few lines of code, which makes it a great way into computer vision for robots, games and smart gadgets.

Course Content

1. Installing CVZone - set up a Python virtual environment with OpenCV, MediaPipe and CVZone, and connect a camera
2. Face detection - find faces in a live video feed
3. Hand detection - track hands and count raised fingers
4. Pose detection - find 33 body landmarks and measure angles between them
5. Annotating your video - add live text and feedback to the picture

Key Results

After completing this course, you will be able to:

• Set up a Raspberry Pi for computer vision with Python
• Capture video from a USB webcam or a Raspberry Pi camera module
• Detect faces, hands and body poses with CVZone
• Use landmark positions to measure distances and angles
• Overlay live feedback on the video stream

What you’ll need

• A Raspberry Pi 4 or 5 running the 64-bit version of Raspberry Pi OS, with a desktop
• A USB webcam or a Raspberry Pi camera module
• A keyboard, mouse and screen, or remote desktop access (the examples open a video window)

The Raspberry Pi 5 gives the smoothest frame rates, but a Pi 4 works well too.

Prerequisites

You should be comfortable writing and running simple Python programs. If you’re new to Python, start with Python for beginners.

How the course works

Each lesson builds a short, complete program you can run straight away, then explains how it works. Most lessons end with a quick quiz, and your progress is saved in your browser and shown in the sidebar.

Next >

You can use the arrows  ← → on your keyboard to navigate between lessons.