Thread by @GantLaborde

Quick Tweet Storm ⛈

How does AI bounding box detection work?

🧠 Learn in 30 seconds

#100DaysOfCode #CodeNewbie #MadeWithTFJS #MachineLearning #ComputerVision

It looks so simple when #AI does it right?

But #machinelearning doesn't give you an image, it gives you data. It's up to you to make it look simple.

You might think a #FrontEnd box gives you four values, and you're right, but it only gives you TWO points. From that you can infer a box to draw with #html5.

However... it very rarely gives you one box. Lots of times you get a whole bunch of boxes!

This is because the #NeuralNetwork is often trained to return X results no matter what. They just adjust the confidence.

Each of these boxes will have a score of confidence, and it's your job to identify if they are identifying the same object, or separate objects for proper #DataVisualization.

There's some well known algorithms to take the best of the best from overlapping results.

Once you have the boxes you're interested in, you can draw them easily with a canvas element.

In #JavaScript you just overlay a canvas on the image and draw rectangles there.

The final result lets you draw clean object detection on #Web.

Now you have an amazing website that can use #ML to detect objects in photos and video!

Want the code to do this? It's all in Chapter 6 of my @OReillyMedia book with a deep explanation.

If you know JavaScript, you can implement @TensorFlow

Learn more about the book here: infinite.red/learning-tenso…

Looking forward to the cool things you decide to make!

Start your #100DaysOfML now