Build a Teachable Machine Image Classifier With Your Kid

Introduce your child to the fundamentals of machine learning through a hands-on, zero-code project.

Children today are surrounded by artificial intelligence, from video recommendation algorithms to voice assistants. However, they rarely understand how these systems make decisions. Instead of letting your children remain passive consumers of technology, you can guide them to become active creators. Using Google's free Teachable Machine tool, you and your child (ages 8 to 16) can build, train, and test a custom computer vision model in under twenty minutes. This hands-on project requires no coding, no paid subscriptions, and no specialized hardware, making it the perfect weekend activity to demystify artificial intelligence.

Moving From Consumption to Creation

For busy Executives, finding high-quality, educational activities to share with your children can be challenging. Many STEM kits require complex setups, expensive subscriptions, or hours of troubleshooting. This project is different. It leverages your computer's built-in webcam and web browser to demonstrate the core mechanics of machine learning. By building an image classifier, your child will learn that AI is not magic. Instead, it is a system that learns from patterns in data. This experience shifts their perspective from viewing AI as an all-knowing entity to understanding it as a tool built and trained by humans.

Why Computer Vision Matters

Computer vision is the field of artificial intelligence that enables computers to interpret and understand the visual world. From facial recognition on smartphones to obstacle detection in autonomous vehicles, computer vision is a cornerstone of modern technology. By building a simple classifier, your child gains firsthand experience with how these complex systems operate. They will see how a computer breaks down an image into pixels, identifies patterns, and associates those patterns with specific labels. This foundational understanding prepares them for a future where AI literacy is as essential as basic computer literacy.

Step 1: Choosing Your Classification Project

Before opening the software, sit down with your child and decide what you want your AI model to recognize. The best projects involve distinct, easily accessible objects. Here are three highly engaging ideas:

• Rock, Paper, Scissors: Train the model to recognize hand gestures. This is an excellent way to show how AI can interpret human movement.
• Healthy vs. Fun Snacks: Use an apple and a cookie. This project can spark interesting conversations about nutrition while teaching computer vision.
• Toy Classifier: Train the model to distinguish between a Lego brick and a toy car. This is ideal for younger children who want to use their favorite toys.

Step 2: Setting Up the Workspace

Open a web browser on your laptop or desktop computer and navigate directly to https://teachablemachine.withgoogle.com/ . Click on the "Get Started" button, and then select "Image Project" followed by "Standard Image Model". This will open the training interface, which is divided into three clear columns: Gather Data, Train Model, and Preview.

Step 3: Gathering and Labeling Training Data

In the first column, you will see two default classes labeled "Class 1" and "Class 2". Have your child rename these classes to match your chosen objects (for example, "Lego" and "Toy Car").

Click the webcam button under the first class. Have your child hold the first object in front of the camera. Click and hold the "Record" button to capture at least one hundred images. Encourage your child to rotate the object, move it closer and further away, and tilt it. This variety helps the model learn the object's overall shape rather than just one specific angle. Repeat this exact process for the second class using the second object.

Step 4: Training the Model

Once both classes have sufficient images, move to the middle column and click "Train Model". This process takes about ten to thirty seconds. Explain to your child that the computer is currently looking at all the images, finding patterns, and learning what makes a Lego brick look different from a toy car. Make sure to keep the browser tab open and active during this brief training phase.

Step 5: Testing and Debugging

Once training is complete, the third column will show a live preview from your webcam. Have your child hold up one of the objects. The model will display a real-time percentage indicating how confident it is about which object it sees.

This is where the real learning happens. Try holding up a completely different object, or holding the original object in a different light. If the model makes a mistake, explain that this is called a false positive. Have your child analyze why the model got confused. Did it focus on the color of your shirt instead of the object? You can easily fix this by adding more diverse images to your classes and retraining the model.

Parent & Educator Sidebar

Core AI Concept: Supervised Learning Supervised learning is the process of training an AI model using labeled data. In this project, the labels are the names of the classes (like "Lego"), and the data consists of the webcam images. The computer learns by finding mathematical patterns that distinguish one labeled group from another.

Conversation Starters:

• "How do you think the computer knows the difference between these two objects when it cannot actually feel or understand them?"
• "What would happen if we only trained the model with red Legos, and then showed it a blue Lego? Why would it get confused?"
• "How do you think self-driving cars use this exact technology to recognize stop signs and pedestrians on the road?"

Action Steps Summary

• Choose Your Objects: Select two distinct items to classify, such as a toy car and a Lego brick.
• Open Teachable Machine: Go to https://teachablemachine.withgoogle.com/ and start an Image Project.
• Capture Training Images: Record at least one hundred webcam images for each object from various angles.
• Train Your Model: Click the train button and wait for the browser to process the data.
• Test and Refine: Use the live preview to test the model, and discuss how the AI makes its decisions.

Looking for more engaging, family-friendly AI projects to build together?