a visual course

From Pixels
to Predictions

(Computer Vision & CNNs)

Understand neural network foundations in less than an hour.

A free, visual course on how AI sees images — from a single pixel to a trained convolutional neural network (CNN). Understand the foundations of computer vision, neural networks, and deep learning in about an hour.

in the real world

The same ideas read medical scans, unlock your phone with your face, let self-driving cars see the road, and power the photo search in your pocket.

Start from the first pixel →

Part 1 — An image is just numbers

1

Pixels & the grid

Zoom into a photo until it dissolves into a grid of numbers.

2

Color & channels

Grayscale 0–255, then RGB as three stacked layers.

3

Images as tensors

Height × width × channels — a picture as a stack of matrices.

Part 2 — Doing things to pixels

4

Pixel math

Brightness, contrast, threshold — functions over every pixel.

5

Convolution & filters

A kernel slides across the grid. The keystone idea.

6

Features: edges & gradients

Why edges carry more signal than raw pixels.

Part 3 — Why this is hard for machines

7

The curse of raw pixels

120,000 numbers, and a 5px shift breaks everything.

8

Hand-designed vs learned

From features we design to features a model discovers.

Part 4 — Teaching a machine to see

9

A neuron & a tiny network

Weights × inputs + bias → a prediction.

10

Convolutional networks

The sliding kernel — but now the numbers are learned.

11

How a CNN sees

Edges → textures → parts → objects, layer by layer.

12

Training: how it learns

Loss, gradient descent, and the ball rolling downhill.

Part 5 — Out in the world

13

Where this shows up

The same pixels-to-predictions pipeline, quietly running in medicine, phones, cars and more.