In simple words: Pixels are numbers; vision models learn patterns across pixels.
What a computer sees
Computer Vision (CV) uses algorithms to interpret images or video. A digital image is a grid of pixels. Each pixel stores numeric colour or brightness values. A colour image commonly has red, green, and blue channels.
From pixels to meaning
A single pixel says little about whether a picture contains a cat. Useful patterns arise from nearby pixels and larger shapes. Convolutional Neural Networks (CNNs) were designed to learn these spatial patterns.
See the flow
pixel grid→local patterns→object clues
Simple example
A 100 by 100 colour image has 10,000 pixel positions, each with colour values. A model looks for patterns across them.
Remember: Pixels are numbers; vision models learn patterns across pixels.